fix: isolate extra unet paths from checkpoints to prevent type misclassification

Refactor _prepare_checkpoint_paths() to return a tuple instead of having side effects on instance variables. This prevents extra unet paths from being incorrectly classified as checkpoints when processing extra paths. - Changed return type from List[str] to Tuple[List[str], List[str], List[str]] (all_paths, checkpoint_roots, unet_roots) - Updated _init_checkpoint_paths() and _apply_library_paths() callers - Fixed extra paths processing to properly isolate main and extra roots - Updated test_checkpoint_path_overlap.py tests for new API This ensures models in extra unet paths are correctly identified as diffusion_model type and don't appear in checkpoints list.
fix(services): implement stable sorting for model and recipe caches
2026-03-21 21:22:11 -03:00 · 2026-03-17 22:03:57 +08:00 · 2026-03-17 14:20:23 +08:00 · 2026-03-17 11:24:59 +08:00 · 2026-03-17 11:01:41 +08:00 · 2026-03-17 09:57:27 +08:00
72 changed files with 10298 additions and 1490 deletions
--- a/.docs/batch-import-design.md
+++ b/.docs/batch-import-design.md
@@ -0,0 +1,153 @@
+# Recipe Batch Import Feature Design
+
+## Overview
+Enable users to import multiple images as recipes in a single operation, rather than processing them individually. This feature addresses the need for efficient bulk recipe creation from existing image collections.
+
+## Architecture
+
+```
+┌─────────────────────────────────────────────────────────────────┐
+│                        Frontend                                  │
+├─────────────────────────────────────────────────────────────────┤
+│  BatchImportManager.js                                          │
+│  ├── InputCollector (收集URL列表/目录路径)                        │
+│  ├── ConcurrencyController (自适应并发控制)                       │
+│  ├── ProgressTracker (进度追踪)                                   │
+│  └── ResultAggregator (结果汇总)                                  │
+├─────────────────────────────────────────────────────────────────┤
+│  batch_import_modal.html                                        │
+│  └── 批量导入UI组件                                              │
+├─────────────────────────────────────────────────────────────────┤
+│  batch_import_progress.css                                      │
+│  └── 进度显示样式                                                │
+└─────────────────────────────────────────────────────────────────┘
+                              │
+                              ▼
+┌─────────────────────────────────────────────────────────────────┐
+│                        Backend                                   │
+├─────────────────────────────────────────────────────────────────┤
+│  py/routes/handlers/recipe_handlers.py                          │
+│  ├── start_batch_import() - 启动批量导入                         │
+│  ├── get_batch_import_progress() - 查询进度                      │
+│  └── cancel_batch_import() - 取消导入                            │
+├─────────────────────────────────────────────────────────────────┤
+│  py/services/batch_import_service.py                            │
+│  ├── 自适应并发执行                                              │
+│  ├── 结果汇总                                                    │
+│  └── WebSocket进度广播                                           │
+└─────────────────────────────────────────────────────────────────┘
+```
+
+## API Endpoints
+
+| 端点 | 方法 | 说明 |
+|------|------|------|
+| `/api/lm/recipes/batch-import/start` | POST | 启动批量导入，返回 operation_id |
+| `/api/lm/recipes/batch-import/progress` | GET | 查询进度状态 |
+| `/api/lm/recipes/batch-import/cancel` | POST | 取消导入 |
+
+## Backend Implementation Details
+
+### BatchImportService
+
+Location: `py/services/batch_import_service.py`
+
+Key classes:
+- `BatchImportItem`: Dataclass for individual import item
+- `BatchImportProgress`: Dataclass for tracking progress
+- `BatchImportService`: Main service class
+
+Features:
+- Adaptive concurrency control (adjusts based on success/failure rate)
+- WebSocket progress broadcasting
+- Graceful error handling (individual failures don't stop the batch)
+- Result aggregation
+
+### WebSocket Message Format
+
+```json
+{
+    "type": "batch_import_progress",
+    "operation_id": "xxx",
+    "total": 50,
+    "completed": 23,
+    "success": 21,
+    "failed": 2,
+    "skipped": 0,
+    "current_item": "image_024.png",
+    "status": "running"
+}
+```
+
+### Input Types
+
+1. **URL List**: Array of URLs (http/https)
+2. **Local Paths**: Array of local file paths
+3. **Directory**: Path to directory with optional recursive flag
+
+### Error Handling
+
+- Invalid URLs/paths: Skip and record error
+- Download failures: Record error, continue
+- Metadata extraction failures: Mark as "no metadata"
+- Duplicate detection: Option to skip duplicates
+
+## Frontend Implementation Details (TODO)
+
+### UI Components
+
+1. **BatchImportModal**: Main modal with tabs for URLs/Directory input
+2. **ProgressDisplay**: Real-time progress bar and status
+3. **ResultsSummary**: Final results with success/failure breakdown
+
+### Adaptive Concurrency Controller
+
+```javascript
+class AdaptiveConcurrencyController {
+    constructor(options = {}) {
+        this.minConcurrency = options.minConcurrency || 1;
+        this.maxConcurrency = options.maxConcurrency || 5;
+        this.currentConcurrency = options.initialConcurrency || 3;
+    }
+
+    adjustConcurrency(taskDuration, success) {
+        if (success && taskDuration < 1000 && this.currentConcurrency < this.maxConcurrency) {
+            this.currentConcurrency = Math.min(this.currentConcurrency + 1, this.maxConcurrency);
+        }
+        if (!success || taskDuration > 10000) {
+            this.currentConcurrency = Math.max(this.currentConcurrency - 1, this.minConcurrency);
+        }
+        return this.currentConcurrency;
+    }
+}
+```
+
+## File Structure
+
+```
+Backend (implemented):
+├── py/services/batch_import_service.py            # 后端服务
+├── py/routes/handlers/batch_import_handler.py     # API处理器 (added to recipe_handlers.py)
+├── tests/services/test_batch_import_service.py    # 单元测试
+└── tests/routes/test_batch_import_routes.py       # API集成测试
+
+Frontend (TODO):
+├── static/js/managers/BatchImportManager.js       # 主管理器
+├── static/js/managers/batch/                       # 子模块
+│   ├── ConcurrencyController.js                    # 并发控制
+│   ├── ProgressTracker.js                          # 进度追踪
+│   └── ResultAggregator.js                         # 结果汇总
+├── static/css/components/batch-import-modal.css   # 样式
+└── templates/components/batch_import_modal.html    # Modal模板
+```
+
+## Implementation Status
+
+- [x] Backend BatchImportService
+- [x] Backend API handlers
+- [x] WebSocket progress broadcasting
+- [x] Unit tests
+- [x] Integration tests
+- [ ] Frontend BatchImportManager
+- [ ] Frontend UI components
+- [ ] E2E tests
--- a/.specs/metadata.schema.json
+++ b/.specs/metadata.schema.json
@@ -0,0 +1,464 @@
+{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "$id": "https://github.com/willmiao/ComfyUI-Lora-Manager/.specs/metadata.schema.json",
+  "title": "ComfyUI LoRa Manager Model Metadata",
+  "description": "Schema for .metadata.json sidecar files used by ComfyUI LoRa Manager",
+  "type": "object",
+  "oneOf": [
+    {
+      "title": "LoRA Model Metadata",
+      "properties": {
+        "file_name": {
+          "type": "string",
+          "description": "Filename without extension"
+        },
+        "model_name": {
+          "type": "string",
+          "description": "Display name of the model"
+        },
+        "file_path": {
+          "type": "string",
+          "description": "Full absolute path to the model file"
+        },
+        "size": {
+          "type": "integer",
+          "minimum": 0,
+          "description": "File size in bytes at time of import/download"
+        },
+        "modified": {
+          "type": "number",
+          "description": "Unix timestamp when model was imported/added (Date Added)"
+        },
+        "sha256": {
+          "type": "string",
+          "pattern": "^[a-f0-9]{64}$",
+          "description": "SHA256 hash of the model file (lowercase)"
+        },
+        "base_model": {
+          "type": "string",
+          "description": "Base model type (SD1.5, SD2.1, SDXL, SD3, Flux, Unknown, etc.)"
+        },
+        "preview_url": {
+          "type": "string",
+          "description": "Path to preview image file"
+        },
+        "preview_nsfw_level": {
+          "type": "integer",
+          "minimum": 0,
+          "default": 0,
+          "description": "NSFW level using bitmask values: 0 (none), 1 (PG), 2 (PG13), 4 (R), 8 (X), 16 (XXX), 32 (Blocked)"
+        },
+        "notes": {
+          "type": "string",
+          "default": "",
+          "description": "User-defined notes"
+        },
+        "from_civitai": {
+          "type": "boolean",
+          "default": true,
+          "description": "Whether the model originated from Civitai"
+        },
+        "civitai": {
+          "$ref": "#/definitions/civitaiObject"
+        },
+        "tags": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "default": [],
+          "description": "Model tags"
+        },
+        "modelDescription": {
+          "type": "string",
+          "default": "",
+          "description": "Full model description"
+        },
+        "civitai_deleted": {
+          "type": "boolean",
+          "default": false,
+          "description": "Whether the model was deleted from Civitai"
+        },
+        "favorite": {
+          "type": "boolean",
+          "default": false,
+          "description": "Whether the model is marked as favorite"
+        },
+        "exclude": {
+          "type": "boolean",
+          "default": false,
+          "description": "Whether to exclude from cache/scanning"
+        },
+        "db_checked": {
+          "type": "boolean",
+          "default": false,
+          "description": "Whether checked against archive database"
+        },
+        "skip_metadata_refresh": {
+          "type": "boolean",
+          "default": false,
+          "description": "Skip this model during bulk metadata refresh"
+        },
+        "metadata_source": {
+          "type": ["string", "null"],
+          "enum": ["civitai_api", "civarchive", "archive_db", null],
+          "default": null,
+          "description": "Last provider that supplied metadata"
+        },
+        "last_checked_at": {
+          "type": "number",
+          "default": 0,
+          "description": "Unix timestamp of last metadata check"
+        },
+        "hash_status": {
+          "type": "string",
+          "enum": ["pending", "calculating", "completed", "failed"],
+          "default": "completed",
+          "description": "Hash calculation status"
+        },
+        "usage_tips": {
+          "type": "string",
+          "default": "{}",
+          "description": "JSON string containing recommended usage parameters (LoRA only)"
+        }
+      },
+      "required": [
+        "file_name",
+        "model_name",
+        "file_path",
+        "size",
+        "modified",
+        "sha256",
+        "base_model"
+      ],
+      "additionalProperties": true
+    },
+    {
+      "title": "Checkpoint Model Metadata",
+      "properties": {
+        "file_name": {
+          "type": "string"
+        },
+        "model_name": {
+          "type": "string"
+        },
+        "file_path": {
+          "type": "string"
+        },
+        "size": {
+          "type": "integer",
+          "minimum": 0
+        },
+        "modified": {
+          "type": "number"
+        },
+        "sha256": {
+          "type": "string",
+          "pattern": "^[a-f0-9]{64}$"
+        },
+        "base_model": {
+          "type": "string"
+        },
+        "preview_url": {
+          "type": "string"
+        },
+        "preview_nsfw_level": {
+          "type": "integer",
+          "minimum": 0,
+          "maximum": 3,
+          "default": 0
+        },
+        "notes": {
+          "type": "string",
+          "default": ""
+        },
+        "from_civitai": {
+          "type": "boolean",
+          "default": true
+        },
+        "civitai": {
+          "$ref": "#/definitions/civitaiObject"
+        },
+        "tags": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "default": []
+        },
+        "modelDescription": {
+          "type": "string",
+          "default": ""
+        },
+        "civitai_deleted": {
+          "type": "boolean",
+          "default": false
+        },
+        "favorite": {
+          "type": "boolean",
+          "default": false
+        },
+        "exclude": {
+          "type": "boolean",
+          "default": false
+        },
+        "db_checked": {
+          "type": "boolean",
+          "default": false
+        },
+        "skip_metadata_refresh": {
+          "type": "boolean",
+          "default": false
+        },
+        "metadata_source": {
+          "type": ["string", "null"],
+          "enum": ["civitai_api", "civarchive", "archive_db", null],
+          "default": null
+        },
+        "last_checked_at": {
+          "type": "number",
+          "default": 0
+        },
+        "hash_status": {
+          "type": "string",
+          "enum": ["pending", "calculating", "completed", "failed"],
+          "default": "completed"
+        },
+        "sub_type": {
+          "type": "string",
+          "default": "checkpoint",
+          "description": "Model sub-type (checkpoint, diffusion_model, etc.)"
+        }
+      },
+      "required": [
+        "file_name",
+        "model_name",
+        "file_path",
+        "size",
+        "modified",
+        "sha256",
+        "base_model"
+      ],
+      "additionalProperties": true
+    },
+    {
+      "title": "Embedding Model Metadata",
+      "properties": {
+        "file_name": {
+          "type": "string"
+        },
+        "model_name": {
+          "type": "string"
+        },
+        "file_path": {
+          "type": "string"
+        },
+        "size": {
+          "type": "integer",
+          "minimum": 0
+        },
+        "modified": {
+          "type": "number"
+        },
+        "sha256": {
+          "type": "string",
+          "pattern": "^[a-f0-9]{64}$"
+        },
+        "base_model": {
+          "type": "string"
+        },
+        "preview_url": {
+          "type": "string"
+        },
+        "preview_nsfw_level": {
+          "type": "integer",
+          "minimum": 0,
+          "maximum": 3,
+          "default": 0
+        },
+        "notes": {
+          "type": "string",
+          "default": ""
+        },
+        "from_civitai": {
+          "type": "boolean",
+          "default": true
+        },
+        "civitai": {
+          "$ref": "#/definitions/civitaiObject"
+        },
+        "tags": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "default": []
+        },
+        "modelDescription": {
+          "type": "string",
+          "default": ""
+        },
+        "civitai_deleted": {
+          "type": "boolean",
+          "default": false
+        },
+        "favorite": {
+          "type": "boolean",
+          "default": false
+        },
+        "exclude": {
+          "type": "boolean",
+          "default": false
+        },
+        "db_checked": {
+          "type": "boolean",
+          "default": false
+        },
+        "skip_metadata_refresh": {
+          "type": "boolean",
+          "default": false
+        },
+        "metadata_source": {
+          "type": ["string", "null"],
+          "enum": ["civitai_api", "civarchive", "archive_db", null],
+          "default": null
+        },
+        "last_checked_at": {
+          "type": "number",
+          "default": 0
+        },
+        "hash_status": {
+          "type": "string",
+          "enum": ["pending", "calculating", "completed", "failed"],
+          "default": "completed"
+        },
+        "sub_type": {
+          "type": "string",
+          "default": "embedding",
+          "description": "Model sub-type"
+        }
+      },
+      "required": [
+        "file_name",
+        "model_name",
+        "file_path",
+        "size",
+        "modified",
+        "sha256",
+        "base_model"
+      ],
+      "additionalProperties": true
+    }
+  ],
+  "definitions": {
+    "civitaiObject": {
+      "type": "object",
+      "default": {},
+      "description": "Civitai/CivArchive API data and user-defined fields",
+      "properties": {
+        "id": {
+          "type": "integer",
+          "description": "Version ID from Civitai"
+        },
+        "modelId": {
+          "type": "integer",
+          "description": "Model ID from Civitai"
+        },
+        "name": {
+          "type": "string",
+          "description": "Version name"
+        },
+        "description": {
+          "type": "string",
+          "description": "Version description"
+        },
+        "baseModel": {
+          "type": "string",
+          "description": "Base model type from Civitai"
+        },
+        "type": {
+          "type": "string",
+          "description": "Model type (checkpoint, embedding, etc.)"
+        },
+        "trainedWords": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Trigger words for the model (from API or user-defined)"
+        },
+        "customImages": {
+          "type": "array",
+          "items": {
+            "type": "object"
+          },
+          "description": "Custom example images added by user"
+        },
+        "model": {
+          "type": "object",
+          "properties": {
+            "name": {
+              "type": "string"
+            },
+            "description": {
+              "type": "string"
+            },
+            "tags": {
+              "type": "array",
+              "items": {
+                "type": "string"
+              }
+            }
+          }
+        },
+        "files": {
+          "type": "array",
+          "items": {
+            "type": "object"
+          }
+        },
+        "images": {
+          "type": "array",
+          "items": {
+            "type": "object"
+          }
+        },
+        "creator": {
+          "type": "object"
+        }
+      },
+      "additionalProperties": true
+    },
+    "usageTips": {
+      "type": "object",
+      "description": "Structure for usage_tips JSON string (LoRA models)",
+      "properties": {
+        "strength_min": {
+          "type": "number",
+          "description": "Minimum recommended model strength"
+        },
+        "strength_max": {
+          "type": "number",
+          "description": "Maximum recommended model strength"
+        },
+        "strength_range": {
+          "type": "string",
+          "description": "Human-readable strength range"
+        },
+        "strength": {
+          "type": "number",
+          "description": "Single recommended strength value"
+        },
+        "clip_strength": {
+          "type": "number",
+          "description": "Recommended CLIP/embedding strength"
+        },
+        "clip_skip": {
+          "type": "integer",
+          "description": "Recommended CLIP skip value"
+        }
+      },
+      "additionalProperties": true
+    }
+  }
+}
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -135,7 +135,7 @@ npm run test:coverage       # Generate coverage report
 - ALWAYS use English for comments (per copilot-instructions.md)
 - Dual mode: ComfyUI plugin (folder_paths) vs standalone (settings.json)
 - Detection: `os.environ.get("LORA_MANAGER_STANDALONE", "0") == "1"`
- Run `python scripts/sync_translation_keys.py` after UI string updates
+- Run `python scripts/sync_translation_keys.py` after adding UI strings to `locales/en.json`
 - Symlinks require normalized paths

 ## Frontend UI Architecture
--- a/README.md
+++ b/README.md
@@ -59,6 +59,7 @@ Insomnia Art Designs, megakirbs, Brennok, wackop, 2018cfh, Takkan, stone9k, $Met
 ### v1.0.0
 * **Extra Folder Paths Support** - Added support for additional model root paths exclusive to LoRA Manager. This allows loading LoRAs from extra locations outside ComfyUI's standard folders, helping avoid performance issues when working with large model libraries.
 * **Settings UI Overhaul** - Redesigned the Settings interface with a more organized layout, making it easier to find and configure application settings.
+* **Lazy Hash Computation** - Implemented lazy hash calculation for large model files (checkpoints and diffusion models). Hashes are now computed only when strictly necessary, minimizing redundant disk I/O and significantly accelerating application initialization.
 * **Milestone & Supporter Recognition** - Updated the Supporter window to show appreciation for all project supporters as this v1.0.0 milestone is reached. Great thanks to the community for the ongoing support!
 * **Bug Fixes & UX Enhancements** - Various bug fixes and user experience improvements for a smoother workflow.

@@ -193,7 +194,7 @@ Insomnia Art Designs, megakirbs, Brennok, wackop, 2018cfh, Takkan, stone9k, $Met

 ### Option 2: **Portable Standalone Edition** (No ComfyUI required)

-1. Download the [Portable Package](https://github.com/willmiao/ComfyUI-Lora-Manager/releases/download/v0.9.8/lora_manager_portable.7z)
+1. Download the [Portable Package](https://github.com/willmiao/ComfyUI-Lora-Manager/releases/download/v1.0.0/lora_manager_portable.7z)
 2. Copy the provided `settings.json.example` file to create a new file named `settings.json` in `comfyui-lora-manager` folder.
 3. Edit the new `settings.json` to include your correct model folder paths and CivitAI API key
   - Set `"use_portable_settings": true` if you want the configuration to remain inside the repository folder instead of your user settings directory.
@@ -320,6 +321,12 @@ npm run test:coverage

 ---

+## Documentation
+
+- **[metadata.json Schema Documentation](docs/metadata-json-schema.md)** — Complete reference for the `.metadata.json` sidecar file format, including all fields, types, and examples for LoRA, Checkpoint, and Embedding models.
+
+---
+
 ## Contributing

 Thank you for your interest in contributing to ComfyUI LoRA Manager! As this project is currently in its early stages and undergoing rapid development and refactoring, we are temporarily not accepting pull requests.
--- a/init.py
+++ b/init.py
@@ -1,6 +1,8 @@
 try:  # pragma: no cover - import fallback for pytest collection
    from .py.lora_manager import LoraManager
    from .py.nodes.lora_loader import LoraLoaderLM, LoraTextLoaderLM
+    from .py.nodes.checkpoint_loader import CheckpointLoaderLM
+    from .py.nodes.unet_loader import UNETLoaderLM
    from .py.nodes.trigger_word_toggle import TriggerWordToggleLM
    from .py.nodes.prompt import PromptLM
    from .py.nodes.text import TextLM
@@ -27,12 +29,12 @@ except (
    PromptLM = importlib.import_module("py.nodes.prompt").PromptLM
    TextLM = importlib.import_module("py.nodes.text").TextLM
    LoraManager = importlib.import_module("py.lora_manager").LoraManager
-    LoraLoaderLM = importlib.import_module(
-        "py.nodes.lora_loader"
-    ).LoraLoaderLM
-    LoraTextLoaderLM = importlib.import_module(
-        "py.nodes.lora_loader"
-    ).LoraTextLoaderLM
+    LoraLoaderLM = importlib.import_module("py.nodes.lora_loader").LoraLoaderLM
+    LoraTextLoaderLM = importlib.import_module("py.nodes.lora_loader").LoraTextLoaderLM
+    CheckpointLoaderLM = importlib.import_module(
+        "py.nodes.checkpoint_loader"
+    ).CheckpointLoaderLM
+    UNETLoaderLM = importlib.import_module("py.nodes.unet_loader").UNETLoaderLM
    TriggerWordToggleLM = importlib.import_module(
        "py.nodes.trigger_word_toggle"
    ).TriggerWordToggleLM
@@ -49,9 +51,7 @@ except (
    LoraRandomizerLM = importlib.import_module(
        "py.nodes.lora_randomizer"
    ).LoraRandomizerLM
-    LoraCyclerLM = importlib.import_module(
-        "py.nodes.lora_cycler"
-    ).LoraCyclerLM
+    LoraCyclerLM = importlib.import_module("py.nodes.lora_cycler").LoraCyclerLM
    init_metadata_collector = importlib.import_module("py.metadata_collector").init

 NODE_CLASS_MAPPINGS = {
@@ -59,6 +59,8 @@ NODE_CLASS_MAPPINGS = {
    TextLM.NAME: TextLM,
    LoraLoaderLM.NAME: LoraLoaderLM,
    LoraTextLoaderLM.NAME: LoraTextLoaderLM,
+    CheckpointLoaderLM.NAME: CheckpointLoaderLM,
+    UNETLoaderLM.NAME: UNETLoaderLM,
    TriggerWordToggleLM.NAME: TriggerWordToggleLM,
    LoraStackerLM.NAME: LoraStackerLM,
    SaveImageLM.NAME: SaveImageLM,
--- a/docs/metadata-json-schema.md
+++ b/docs/metadata-json-schema.md
@@ -0,0 +1,363 @@
+# metadata.json Schema Documentation
+
+This document defines the complete schema for `.metadata.json` files used by Lora Manager. These sidecar files store model metadata alongside model files (LoRA, Checkpoint, Embedding).
+
+## Overview
+
+- **File naming**: `<model_name>.metadata.json` (e.g., `my_lora.safetensors` → `my_lora.metadata.json`)
+- **Format**: JSON with UTF-8 encoding
+- **Purpose**: Store model metadata, tags, descriptions, preview images, and Civitai/CivArchive integration data
+- **Extensibility**: Unknown fields are preserved via `_unknown_fields` mechanism for forward compatibility
+
+---
+
+## Base Fields (All Model Types)
+
+These fields are present in all model metadata files.
+
+| Field | Type | Required | Auto-Updated | Description |
+|-------|------|----------|--------------|-------------|
+| `file_name` | string | ✅ Yes | ✅ Yes | Filename without extension (e.g., `"my_lora"`) |
+| `model_name` | string | ✅ Yes | ❌ No | Display name of the model. **Default**: `file_name` if no other source |
+| `file_path` | string | ✅ Yes | ✅ Yes | Full absolute path to the model file (normalized with `/` separators) |
+| `size` | integer | ✅ Yes | ❌ No | File size in bytes. **Set at**: Initial scan or download completion. Does not change thereafter. |
+| `modified` | float | ✅ Yes | ❌ No | **Import timestamp** — Unix timestamp when the model was first imported/added to the system. Used for "Date Added" sorting. Does not change after initial creation. |
+| `sha256` | string | ⚠️ Conditional | ✅ Yes | SHA256 hash of the model file (lowercase). **LoRA**: Required. **Checkpoint**: May be empty when `hash_status="pending"` (lazy hash calculation) |
+| `base_model` | string | ❌ No | ❌ No | Base model type. **Examples**: `"SD 1.5"`, `"SDXL 1.0"`, `"SDXL Lightning"`, `"Flux.1 D"`, `"Flux.1 S"`, `"Flux.1 Krea"`, `"Illustrious"`, `"Pony"`, `"AuraFlow"`, `"Kolors"`, `"ZImageTurbo"`, `"Wan Video"`, etc. **Default**: `"Unknown"` or `""` |
+| `preview_url` | string | ❌ No | ✅ Yes | Path to preview image file |
+| `preview_nsfw_level` | integer | ❌ No | ❌ No | NSFW level using **bitmask values** from Civitai: `1` (PG), `2` (PG13), `4` (R), `8` (X), `16` (XXX), `32` (Blocked). **Default**: `0` (none) |
+| `notes` | string | ❌ No | ❌ No | User-defined notes |
+| `from_civitai` | boolean | ❌ No (default: `true`) | ❌ No | Whether the model originated from Civitai |
+| `civitai` | object | ❌ No | ⚠️ Partial | Civitai/CivArchive API data and user-defined fields |
+| `tags` | array[string] | ❌ No | ⚠️ Partial | Model tags (merged from API and user input) |
+| `modelDescription` | string | ❌ No | ⚠️ Partial | Full model description (from API or user) |
+| `civitai_deleted` | boolean | ❌ No (default: `false`) | ❌ No | Whether the model was deleted from Civitai |
+| `favorite` | boolean | ❌ No (default: `false`) | ❌ No | Whether the model is marked as favorite |
+| `exclude` | boolean | ❌ No (default: `false`) | ❌ No | Whether to exclude from cache/scanning. User can set from `false` to `true` (currently no UI to revert) |
+| `db_checked` | boolean | ❌ No (default: `false`) | ❌ No | Whether checked against archive database |
+| `skip_metadata_refresh` | boolean | ❌ No (default: `false`) | ❌ No | Skip this model during bulk metadata refresh |
+| `metadata_source` | string\|null | ❌ No | ✅ Yes | Last provider that supplied metadata (see below) |
+| `last_checked_at` | float | ❌ No (default: `0`) | ✅ Yes | Unix timestamp of last metadata check |
+| `hash_status` | string | ❌ No (default: `"completed"`) | ✅ Yes | Hash calculation status: `"pending"`, `"calculating"`, `"completed"`, `"failed"` |
+
+---
+
+## Model-Specific Fields
+
+### LoRA Models
+
+LoRA models do not have a `model_type` field in metadata.json. The type is inferred from context or `civitai.type` (e.g., `"LoRA"`, `"LoCon"`, `"DoRA"`).
+
+| Field | Type | Required | Auto-Updated | Description |
+|-------|------|----------|--------------|-------------|
+| `usage_tips` | string (JSON) | ❌ No (default: `"{}"`) | ❌ No | JSON string containing recommended usage parameters |
+
+**`usage_tips` JSON structure:**
+
+```json
+{
+  "strength_min": 0.3,
+  "strength_max": 0.8,
+  "strength_range": "0.3-0.8",
+  "strength": 0.6,
+  "clip_strength": 0.5,
+  "clip_skip": 2
+}
+```
+
+| Key | Type | Description |
+|-----|------|-------------|
+| `strength_min` | number | Minimum recommended model strength |
+| `strength_max` | number | Maximum recommended model strength |
+| `strength_range` | string | Human-readable strength range |
+| `strength` | number | Single recommended strength value |
+| `clip_strength` | number | Recommended CLIP/embedding strength |
+| `clip_skip` | integer | Recommended CLIP skip value |
+
+---
+
+### Checkpoint Models
+
+| Field | Type | Required | Auto-Updated | Description |
+|-------|------|----------|--------------|-------------|
+| `model_type` | string | ❌ No (default: `"checkpoint"`) | ❌ No | Model type: `"checkpoint"`, `"diffusion_model"` |
+
+---
+
+### Embedding Models
+
+| Field | Type | Required | Auto-Updated | Description |
+|-------|------|----------|--------------|-------------|
+| `model_type` | string | ❌ No (default: `"embedding"`) | ❌ No | Model type: `"embedding"` |
+
+---
+
+## The `civitai` Field Structure
+
+The `civitai` object stores the complete Civitai/CivArchive API response. Lora Manager preserves all fields from the API for future compatibility and extracts specific fields for use in the application.
+
+### Version-Level Fields (Civitai API)
+
+**Fields Used by Lora Manager:**
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `id` | integer | Version ID |
+| `modelId` | integer | Parent model ID |
+| `name` | string | Version name (e.g., `"v1.0"`, `"v2.0-pruned"`) |
+| `nsfwLevel` | integer | NSFW level (bitmask: 1=PG, 2=PG13, 4=R, 8=X, 16=XXX, 32=Blocked) |
+| `baseModel` | string | Base model (e.g., `"SDXL 1.0"`, `"Flux.1 D"`, `"Illustrious"`, `"Pony"`) |
+| `trainedWords` | array[string] | **Trigger words** for the model |
+| `type` | string | Model type (`"LoRA"`, `"Checkpoint"`, `"TextualInversion"`) |
+| `earlyAccessEndsAt` | string\|null | Early access end date (used for update notifications) |
+| `description` | string | Version description (HTML) |
+| `model` | object | Parent model object (see Model-Level Fields below) |
+| `creator` | object | Creator information (see Creator Fields below) |
+| `files` | array[object] | File list with hashes, sizes, download URLs (used for metadata extraction) |
+| `images` | array[object] | Image list with metadata, prompts, NSFW levels (used for preview/examples) |
+
+**Fields Stored but Not Currently Used:**
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `createdAt` | string (ISO 8601) | Creation timestamp |
+| `updatedAt` | string (ISO 8601) | Last update timestamp |
+| `status` | string | Version status (e.g., `"Published"`, `"Draft"`) |
+| `publishedAt` | string (ISO 8601) | Publication timestamp |
+| `baseModelType` | string | Base model type (e.g., `"Standard"`, `"Inpaint"`, `"Refiner"`) |
+| `earlyAccessConfig` | object | Early access configuration |
+| `uploadType` | string | Upload type (`"Created"`, `"FineTuned"`, etc.) |
+| `usageControl` | string | Usage control setting |
+| `air` | string | Artifact ID (URN format: `urn:air:sdxl:lora:civitai:122359@135867`) |
+| `stats` | object | Download count, ratings, thumbs up count |
+| `videos` | array[object] | Video list |
+| `downloadUrl` | string | Direct download URL |
+| `trainingStatus` | string\|null | Training status (for on-site training) |
+| `trainingDetails` | object\|null | Training configuration |
+
+### Model-Level Fields (`civitai.model.*`)
+
+**Fields Used by Lora Manager:**
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `name` | string | Model name |
+| `type` | string | Model type (`"LoRA"`, `"Checkpoint"`, `"TextualInversion"`) |
+| `description` | string | Model description (HTML, used for `modelDescription`) |
+| `tags` | array[string] | Model tags (used for `tags` field) |
+| `allowNoCredit` | boolean | License: allow use without credit |
+| `allowCommercialUse` | array[string] | License: allowed commercial uses. **Values**: `"Image"` (sell generated images), `"Video"` (sell generated videos), `"RentCivit"` (rent on Civitai), `"Rent"` (rent elsewhere) |
+| `allowDerivatives` | boolean | License: allow derivatives |
+| `allowDifferentLicense` | boolean | License: allow different license |
+
+**Fields Stored but Not Currently Used:**
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `nsfw` | boolean | Model NSFW flag |
+| `poi` | boolean | Person of Interest flag |
+
+### Creator Fields (`civitai.creator.*`)
+
+Both fields are used by Lora Manager:
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `username` | string | Creator username (used for author display and search) |
+| `image` | string | Creator avatar URL (used for display) |
+
+### Model Type Field (Top-Level, Outside `civitai`)
+
+| Field | Type | Values | Description |
+|-------|------|--------|-------------|
+| `model_type` | string | `"checkpoint"`, `"diffusion_model"`, `"embedding"` | Stored in metadata.json for Checkpoint and Embedding models. **Note**: LoRA models do not have this field; type is inferred from `civitai.type` or context. |
+
+### User-Defined Fields (Within `civitai`)
+
+For models not from Civitai or user-added data:
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `trainedWords` | array[string] | **Trigger words** — manually added by user |
+| `customImages` | array[object] | Custom example images added by user |
+
+### customImages Structure
+
+Each custom image entry has the following structure:
+
+```json
+{
+  "url": "",
+  "id": "short_id",
+  "nsfwLevel": 0,
+  "width": 832,
+  "height": 1216,
+  "type": "image",
+  "meta": {
+    "prompt": "...",
+    "negativePrompt": "...",
+    "steps": 20,
+    "cfgScale": 7,
+    "seed": 123456
+  },
+  "hasMeta": true,
+  "hasPositivePrompt": true
+}
+```
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `url` | string | Empty for local custom images |
+| `id` | string | Short ID or filename |
+| `nsfwLevel` | integer | NSFW level (bitmask) |
+| `width` | integer | Image width in pixels |
+| `height` | integer | Image height in pixels |
+| `type` | string | `"image"` or `"video"` |
+| `meta` | object\|null | Generation metadata (prompt, seed, etc.) extracted from image |
+| `hasMeta` | boolean | Whether metadata is available |
+| `hasPositivePrompt` | boolean | Whether a positive prompt is available |
+
+### Minimal Non-Civitai Example
+
+```json
+{
+  "civitai": {
+    "trainedWords": ["my_trigger_word"]
+  }
+}
+```
+
+### Non-Civitai Example Without Trigger Words
+
+```json
+{
+  "civitai": {}
+}
+```
+
+### Example: User-Added Custom Images
+
+```json
+{
+  "civitai": {
+    "trainedWords": ["custom_style"],
+    "customImages": [
+      {
+        "url": "",
+        "id": "example_1",
+        "nsfwLevel": 0,
+        "width": 832,
+        "height": 1216,
+        "type": "image",
+        "meta": {
+          "prompt": "example prompt",
+          "seed": 12345
+        },
+        "hasMeta": true,
+        "hasPositivePrompt": true
+      }
+    ]
+  }
+}
+```
+
+---
+
+## Metadata Source Values
+
+The `metadata_source` field indicates which provider last updated the metadata:
+
+| Value | Source |
+|-------|--------|
+| `"civitai_api"` | Civitai API |
+| `"civarchive"` | CivArchive API |
+| `"archive_db"` | Metadata Archive Database |
+| `null` | No external source (user-defined only) |
+
+---
+
+## Auto-Update Behavior
+
+### Fields Updated During Scanning
+
+These fields are automatically synchronized with the filesystem:
+
+- `file_name` — Updated if actual filename differs
+- `file_path` — Normalized and updated if path changes
+- `preview_url` — Updated if preview file is moved/removed
+- `sha256` — Updated during hash calculation (when `hash_status="pending"`)
+- `hash_status` — Updated during hash calculation
+- `last_checked_at` — Timestamp of scan
+- `metadata_source` — Set based on metadata provider
+
+### Fields Set Once (Immutable After Import)
+
+These fields are set when the model is first imported/scanned and **never change** thereafter:
+
+- `modified` — Import timestamp (used for "Date Added" sorting)
+- `size` — File size at time of import/download
+
+### User-Editable Fields
+
+These fields can be edited by users at any time through the Lora Manager UI or by manually editing the metadata.json file:
+
+- `model_name` — Display name
+- `tags` — Model tags
+- `modelDescription` — Model description
+- `notes` — User notes
+- `favorite` — Favorite flag
+- `exclude` — Exclude from scanning (user can set `false`→`true`, currently no UI to revert)
+- `skip_metadata_refresh` — Skip during bulk refresh
+- `civitai.trainedWords` — Trigger words
+- `civitai.customImages` — Custom example images
+- `usage_tips` — Usage recommendations (LoRA only)
+
+---
+
+
+## Field Reference by Behavior
+
+### Required Fields (Must Always Exist)
+
+- `file_name`
+- `model_name` (defaults to `file_name` if not provided)
+- `file_path`
+- `size`
+- `modified`
+- `sha256` (LoRA: always required; Checkpoint: may be empty when `hash_status="pending"`)
+
+### Optional Fields with Defaults
+
+| Field | Default |
+|-------|---------|
+| `base_model` | `"Unknown"` or `""` |
+| `preview_nsfw_level` | `0` |
+| `from_civitai` | `true` |
+| `civitai` | `{}` |
+| `tags` | `[]` |
+| `modelDescription` | `""` |
+| `notes` | `""` |
+| `civitai_deleted` | `false` |
+| `favorite` | `false` |
+| `exclude` | `false` |
+| `db_checked` | `false` |
+| `skip_metadata_refresh` | `false` |
+| `metadata_source` | `null` |
+| `last_checked_at` | `0` |
+| `hash_status` | `"completed"` |
+| `usage_tips` | `"{}"` (LoRA only) |
+| `model_type` | `"checkpoint"` or `"embedding"` (not present in LoRA models) |
+
+---
+
+## Version History
+
+| Version | Date | Changes |
+|---------|------|---------|
+| 1.0 | 2026-03 | Initial schema documentation |
+
+---
+
+## See Also
+
+- [JSON Schema Definition](../.specs/metadata.schema.json) — Formal JSON Schema for validation
--- a/locales/de.json
+++ b/locales/de.json
@@ -14,7 +14,8 @@
            "backToTop": "Nach oben",
            "settings": "Einstellungen",
            "help": "Hilfe",
-            "add": "Hinzufügen"
+            "add": "Hinzufügen",
+            "close": "Schließen"
        },
        "status": {
            "loading": "Wird geladen...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "Voreinstellungsname...",
            "baseModel": "Basis-Modell",
            "modelTags": "Tags (Top 20)",
-            "modelTypes": "Model Types",
+            "modelTypes": "Modelltypen",
            "license": "Lizenz",
            "noCreditRequired": "Kein Credit erforderlich",
            "allowSellingGeneratedContent": "Verkauf erlaubt",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "Wenigste"
            },
            "refresh": {
-                "title": "Rezeptliste aktualisieren"
+                "title": "Rezeptliste aktualisieren",
+                "quick": "Änderungen synchronisieren",
+                "quickTooltip": "Änderungen synchronisieren - schnelle Aktualisierung ohne Cache-Neubau",
+                "full": "Cache neu aufbauen",
+                "fullTooltip": "Cache neu aufbauen - vollständiger Rescan aller Rezeptdateien"
            },
            "filteredByLora": "Gefiltert nach LoRA",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "Rezept-Reparatur fehlgeschlagen: {message}",
                "missingId": "Rezept kann nicht repariert werden: Fehlende Rezept-ID"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "Fehler beim Laden der {modelType}s: {message}",
            "refreshComplete": "Aktualisierung abgeschlossen",
            "refreshFailed": "Fehler beim Aktualisieren der Rezepte: {message}",
+            "syncComplete": "Synchronisation abgeschlossen",
+            "syncFailed": "Fehler beim Synchronisieren der Rezepte: {message}",
            "updateFailed": "Fehler beim Aktualisieren des Rezepts: {error}",
            "updateError": "Fehler beim Aktualisieren des Rezepts: {message}",
            "nameSaved": "Rezept \"{name}\" erfolgreich gespeichert",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "Fehler beim Speichern des Rezepts: {error}",
            "importFailed": "Import fehlgeschlagen: {message}",
            "folderTreeFailed": "Fehler beim Laden des Ordnerbaums",
-            "folderTreeError": "Fehler beim Laden des Ordnerbaums"
+            "folderTreeError": "Fehler beim Laden des Ordnerbaums",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "Keine Modelle ausgewählt",
--- a/locales/en.json
+++ b/locales/en.json
@@ -14,7 +14,8 @@
            "backToTop": "Back to top",
            "settings": "Settings",
            "help": "Help",
-            "add": "Add"
+            "add": "Add",
+            "close": "Close"
        },
        "status": {
            "loading": "Loading...",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "Least"
            },
            "refresh": {
-                "title": "Refresh recipe list"
+                "title": "Refresh recipe list",
+                "quick": "Sync Changes",
+                "quickTooltip": "Sync changes - quick refresh without rebuilding cache",
+                "full": "Rebuild Cache",
+                "fullTooltip": "Rebuild cache - full rescan of all recipe files"
            },
            "filteredByLora": "Filtered by LoRA",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "Failed to repair recipe: {message}",
                "missingId": "Cannot repair recipe: Missing recipe ID"
            }
+        },
+        "batchImport": {
+            "title": "Batch Import Recipes",
+            "action": "Batch Import",
+            "urlList": "URL List",
+            "directory": "Directory",
+            "urlDescription": "Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "Enter a directory path to import all images from that folder.",
+            "urlsLabel": "Image URLs or Local Paths",
+            "urlsPlaceholder": "https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "Enter one URL or path per line",
+            "directoryPath": "Directory Path",
+            "directoryPlaceholder": "/path/to/images/folder",
+            "browse": "Browse",
+            "recursive": "Include subdirectories",
+            "tagsOptional": "Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "Enter tags separated by commas",
+            "tagsHint": "Tags will be added to all imported recipes",
+            "skipNoMetadata": "Skip images without metadata",
+            "skipNoMetadataHelp": "Images without LoRA metadata will be skipped automatically.",
+            "start": "Start Import",
+            "startImport": "Start Import",
+            "importing": "Importing...",
+            "progress": "Progress",
+            "total": "Total",
+            "success": "Success",
+            "failed": "Failed",
+            "skipped": "Skipped",
+            "current": "Current",
+            "currentItem": "Current",
+            "preparing": "Preparing...",
+            "cancel": "Cancel",
+            "cancelImport": "Cancel",
+            "cancelled": "Import cancelled",
+            "completed": "Import completed",
+            "completedWithErrors": "Completed with errors",
+            "completedSuccess": "Successfully imported {count} recipe(s)",
+            "successCount": "Successful",
+            "failedCount": "Failed",
+            "skippedCount": "Skipped",
+            "totalProcessed": "Total processed",
+            "viewDetails": "View Details",
+            "newImport": "New Import",
+            "manualPathEntry": "Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "Directory selected: {path}",
+            "batchImportManualEntryRequired": "File browser not available. Please enter the directory path manually.",
+            "backToParent": "Back to parent directory",
+            "folders": "Folders",
+            "folderCount": "{count} folders",
+            "imageFiles": "Image Files",
+            "images": "images",
+            "imageCount": "{count} images",
+            "selectFolder": "Select This Folder",
+            "errors": {
+                "enterUrls": "Please enter at least one URL or path",
+                "enterDirectory": "Please enter a directory path",
+                "startFailed": "Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "Failed to load {modelType}s: {message}",
            "refreshComplete": "Refresh complete",
            "refreshFailed": "Failed to refresh recipes: {message}",
+            "syncComplete": "Sync complete",
+            "syncFailed": "Failed to sync recipes: {message}",
            "updateFailed": "Failed to update recipe: {error}",
            "updateError": "Error updating recipe: {message}",
            "nameSaved": "Recipe \"{name}\" saved successfully",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "Failed to save recipe: {error}",
            "importFailed": "Import failed: {message}",
            "folderTreeFailed": "Failed to load folder tree",
-            "folderTreeError": "Error loading folder tree"
+            "folderTreeError": "Error loading folder tree",
+            "batchImportFailed": "Failed to start batch import: {message}",
+            "batchImportCancelling": "Cancelling batch import...",
+            "batchImportCancelFailed": "Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "Please enter at least one URL or file path",
+            "batchImportNoDirectory": "Please enter a directory path",
+            "batchImportBrowseFailed": "Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "No models selected",
--- a/locales/es.json
+++ b/locales/es.json
@@ -14,7 +14,8 @@
            "backToTop": "Volver arriba",
            "settings": "Configuración",
            "help": "Ayuda",
-            "add": "Añadir"
+            "add": "Añadir",
+            "close": "Cerrar"
        },
        "status": {
            "loading": "Cargando...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "Nombre del preajuste...",
            "baseModel": "Modelo base",
            "modelTags": "Etiquetas (Top 20)",
-            "modelTypes": "Model Types",
+            "modelTypes": "Tipos de modelos",
            "license": "Licencia",
            "noCreditRequired": "Sin crédito requerido",
            "allowSellingGeneratedContent": "Venta permitida",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "Menos"
            },
            "refresh": {
-                "title": "Actualizar lista de recetas"
+                "title": "Actualizar lista de recetas",
+                "quick": "Sincronizar cambios",
+                "quickTooltip": "Sincronizar cambios - actualización rápida sin reconstruir caché",
+                "full": "Reconstruir caché",
+                "fullTooltip": "Reconstruir caché - reescaneo completo de todos los archivos de recetas"
            },
            "filteredByLora": "Filtrado por LoRA",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "Error al reparar la receta: {message}",
                "missingId": "No se puede reparar la receta: falta el ID de la receta"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "Error al cargar {modelType}s: {message}",
            "refreshComplete": "Actualización completa",
            "refreshFailed": "Error al actualizar recetas: {message}",
+            "syncComplete": "Sincronización completa",
+            "syncFailed": "Error al sincronizar recetas: {message}",
            "updateFailed": "Error al actualizar receta: {error}",
            "updateError": "Error actualizando receta: {message}",
            "nameSaved": "Receta \"{name}\" guardada exitosamente",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "Error al guardar receta: {error}",
            "importFailed": "Importación falló: {message}",
            "folderTreeFailed": "Error al cargar árbol de carpetas",
-            "folderTreeError": "Error cargando árbol de carpetas"
+            "folderTreeError": "Error cargando árbol de carpetas",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "No hay modelos seleccionados",
--- a/locales/fr.json
+++ b/locales/fr.json
@@ -14,7 +14,8 @@
            "backToTop": "Retour en haut",
            "settings": "Paramètres",
            "help": "Aide",
-            "add": "Ajouter"
+            "add": "Ajouter",
+            "close": "Fermer"
        },
        "status": {
            "loading": "Chargement...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "Nom du préréglage...",
            "baseModel": "Modèle de base",
            "modelTags": "Tags (Top 20)",
-            "modelTypes": "Model Types",
+            "modelTypes": "Types de modèles",
            "license": "Licence",
            "noCreditRequired": "Crédit non requis",
            "allowSellingGeneratedContent": "Vente autorisée",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "Moins"
            },
            "refresh": {
-                "title": "Actualiser la liste des recipes"
+                "title": "Actualiser la liste des recipes",
+                "quick": "Synchroniser les changements",
+                "quickTooltip": "Synchroniser les changements - actualisation rapide sans reconstruire le cache",
+                "full": "Reconstruire le cache",
+                "fullTooltip": "Reconstruire le cache - rescan complet de tous les fichiers de recipes"
            },
            "filteredByLora": "Filtré par LoRA",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "Échec de la réparation de la recette : {message}",
                "missingId": "Impossible de réparer la recette : ID de recette manquant"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "Échec du chargement des {modelType}s : {message}",
            "refreshComplete": "Actualisation terminée",
            "refreshFailed": "Échec de l'actualisation des recipes : {message}",
+            "syncComplete": "Synchronisation terminée",
+            "syncFailed": "Échec de la synchronisation des recipes : {message}",
            "updateFailed": "Échec de la mise à jour de la recipe : {error}",
            "updateError": "Erreur lors de la mise à jour de la recipe : {message}",
            "nameSaved": "Recipe \"{name}\" sauvegardée avec succès",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "Échec de la sauvegarde de la recipe : {error}",
            "importFailed": "Échec de l'importation : {message}",
            "folderTreeFailed": "Échec du chargement de l'arborescence des dossiers",
-            "folderTreeError": "Erreur lors du chargement de l'arborescence des dossiers"
+            "folderTreeError": "Erreur lors du chargement de l'arborescence des dossiers",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "Aucun modèle sélectionné",
--- a/locales/he.json
+++ b/locales/he.json
@@ -14,7 +14,8 @@
            "backToTop": "חזרה למעלה",
            "settings": "הגדרות",
            "help": "עזרה",
-            "add": "הוספה"
+            "add": "הוספה",
+            "close": "סגור"
        },
        "status": {
            "loading": "טוען...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "שם קביעה מראש...",
            "baseModel": "מודל בסיס",
            "modelTags": "תגיות (20 המובילות)",
-            "modelTypes": "Model Types",
+            "modelTypes": "סוגי מודלים",
            "license": "רישיון",
            "noCreditRequired": "ללא קרדיט נדרש",
            "allowSellingGeneratedContent": "אפשר מכירה",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "הכי פחות"
            },
            "refresh": {
-                "title": "רענן רשימת מתכונים"
+                "title": "רענן רשימת מתכונים",
+                "quick": "סנכרן שינויים",
+                "quickTooltip": "סנכרן שינויים - רענון מהיר ללא בניית מטמון מחדש",
+                "full": "בנה מטמון מחדש",
+                "fullTooltip": "בנה מטמון מחדש - סריקה מחדש מלאה של כל קבצי המתכונים"
            },
            "filteredByLora": "מסונן לפי LoRA",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "תיקון המתכון נכשל: {message}",
                "missingId": "לא ניתן לתקן את המתכון: חסר מזהה מתכון"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "טעינת {modelType}s נכשלה: {message}",
            "refreshComplete": "הרענון הושלם",
            "refreshFailed": "רענון המתכונים נכשל: {message}",
+            "syncComplete": "הסנכרון הושלם",
+            "syncFailed": "סנכרון המתכונים נכשל: {message}",
            "updateFailed": "עדכון המתכון נכשל: {error}",
            "updateError": "שגיאה בעדכון המתכון: {message}",
            "nameSaved": "המתכון \"{name}\" נשמר בהצלחה",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "שמירת המתכון נכשלה: {error}",
            "importFailed": "הייבוא נכשל: {message}",
            "folderTreeFailed": "טעינת עץ התיקיות נכשלה",
-            "folderTreeError": "שגיאה בטעינת עץ התיקיות"
+            "folderTreeError": "שגיאה בטעינת עץ התיקיות",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "לא נבחרו מודלים",
--- a/locales/ja.json
+++ b/locales/ja.json
@@ -14,7 +14,8 @@
            "backToTop": "トップへ戻る",
            "settings": "設定",
            "help": "ヘルプ",
-            "add": "追加"
+            "add": "追加",
+            "close": "閉じる"
        },
        "status": {
            "loading": "読み込み中...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "プリセット名...",
            "baseModel": "ベースモデル",
            "modelTags": "タグ（上位20）",
-            "modelTypes": "Model Types",
+            "modelTypes": "モデルタイプ",
            "license": "ライセンス",
            "noCreditRequired": "クレジット不要",
            "allowSellingGeneratedContent": "販売許可",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "少ない順"
            },
            "refresh": {
-                "title": "レシピリストを更新"
+                "title": "レシピリストを更新",
+                "quick": "変更を同期",
+                "quickTooltip": "変更を同期 - キャッシュを再構築せずにクイック更新",
+                "full": "キャッシュを再構築",
+                "fullTooltip": "キャッシュを再構築 - すべてのレシピファイルを完全に再スキャン"
            },
            "filteredByLora": "LoRAでフィルタ済み",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "レシピの修復に失敗しました: {message}",
                "missingId": "レシピを修復できません: レシピIDがありません"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "{modelType}の読み込みに失敗しました：{message}",
            "refreshComplete": "更新完了",
            "refreshFailed": "レシピの更新に失敗しました：{message}",
+            "syncComplete": "同期完了",
+            "syncFailed": "レシピの同期に失敗しました：{message}",
            "updateFailed": "レシピの更新に失敗しました：{error}",
            "updateError": "レシピ更新エラー：{message}",
            "nameSaved": "レシピ\"{name}\"が正常に保存されました",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "レシピの保存に失敗しました：{error}",
            "importFailed": "インポートに失敗しました：{message}",
            "folderTreeFailed": "フォルダツリーの読み込みに失敗しました",
-            "folderTreeError": "フォルダツリー読み込みエラー"
+            "folderTreeError": "フォルダツリー読み込みエラー",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "モデルが選択されていません",
--- a/locales/ko.json
+++ b/locales/ko.json
@@ -14,7 +14,8 @@
            "backToTop": "맨 위로",
            "settings": "설정",
            "help": "도움말",
-            "add": "추가"
+            "add": "추가",
+            "close": "닫기"
        },
        "status": {
            "loading": "로딩 중...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "프리셋 이름...",
            "baseModel": "베이스 모델",
            "modelTags": "태그 (상위 20개)",
-            "modelTypes": "Model Types",
+            "modelTypes": "모델 유형",
            "license": "라이선스",
            "noCreditRequired": "크레딧 표기 없음",
            "allowSellingGeneratedContent": "판매 허용",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "적은순"
            },
            "refresh": {
-                "title": "레시피 목록 새로고침"
+                "title": "레시피 목록 새로고침",
+                "quick": "변경 사항 동기화",
+                "quickTooltip": "변경 사항 동기화 - 캐시를 재구성하지 않고 빠른 새로고침",
+                "full": "캐시 재구성",
+                "fullTooltip": "캐시 재구성 - 모든 레시피 파일을 완전히 다시 스캔"
            },
            "filteredByLora": "LoRA로 필터링됨",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "레시피 복구 실패: {message}",
                "missingId": "레시피를 복구할 수 없음: 레시피 ID 누락"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "{modelType} 로딩 실패: {message}",
            "refreshComplete": "새로고침 완료",
            "refreshFailed": "레시피 새로고침 실패: {message}",
+            "syncComplete": "동기화 완료",
+            "syncFailed": "레시피 동기화 실패: {message}",
            "updateFailed": "레시피 업데이트 실패: {error}",
            "updateError": "레시피 업데이트 오류: {message}",
            "nameSaved": "레시피 \"{name}\"이 성공적으로 저장되었습니다",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "레시피 저장 실패: {error}",
            "importFailed": "가져오기 실패: {message}",
            "folderTreeFailed": "폴더 트리 로딩 실패",
-            "folderTreeError": "폴더 트리 로딩 오류"
+            "folderTreeError": "폴더 트리 로딩 오류",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "선택된 모델이 없습니다",
--- a/locales/ru.json
+++ b/locales/ru.json
@@ -14,7 +14,8 @@
            "backToTop": "Наверх",
            "settings": "Настройки",
            "help": "Справка",
-            "add": "Добавить"
+            "add": "Добавить",
+            "close": "Закрыть"
        },
        "status": {
            "loading": "Загрузка...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "Имя пресета...",
            "baseModel": "Базовая модель",
            "modelTags": "Теги (Топ 20)",
-            "modelTypes": "Model Types",
+            "modelTypes": "Типы моделей",
            "license": "Лицензия",
            "noCreditRequired": "Без указания авторства",
            "allowSellingGeneratedContent": "Продажа разрешена",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "Меньше всего"
            },
            "refresh": {
-                "title": "Обновить список рецептов"
+                "title": "Обновить список рецептов",
+                "quick": "Синхронизировать изменения",
+                "quickTooltip": "Синхронизировать изменения - быстрое обновление без перестроения кэша",
+                "full": "Перестроить кэш",
+                "fullTooltip": "Перестроить кэш - полное повторное сканирование всех файлов рецептов"
            },
            "filteredByLora": "Фильтр по LoRA",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "Не удалось восстановить рецепт: {message}",
                "missingId": "Не удалось восстановить рецепт: отсутствует ID рецепта"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "Не удалось загрузить {modelType}s: {message}",
            "refreshComplete": "Обновление завершено",
            "refreshFailed": "Не удалось обновить рецепты: {message}",
+            "syncComplete": "Синхронизация завершена",
+            "syncFailed": "Не удалось синхронизировать рецепты: {message}",
            "updateFailed": "Не удалось обновить рецепт: {error}",
            "updateError": "Ошибка обновления рецепта: {message}",
            "nameSaved": "Рецепт \"{name}\" успешно сохранен",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "Не удалось сохранить рецепт: {error}",
            "importFailed": "Импорт не удался: {message}",
            "folderTreeFailed": "Не удалось загрузить дерево папок",
-            "folderTreeError": "Ошибка загрузки дерева папок"
+            "folderTreeError": "Ошибка загрузки дерева папок",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "Модели не выбраны",
--- a/locales/zh-CN.json
+++ b/locales/zh-CN.json
@@ -14,7 +14,8 @@
            "backToTop": "返回顶部",
            "settings": "设置",
            "help": "帮助",
-            "add": "添加"
+            "add": "添加",
+            "close": "关闭"
        },
        "status": {
            "loading": "加载中...",
@@ -162,11 +163,11 @@
            "error": "清理示例图片文件夹失败：{message}"
        },
        "fetchMissingLicenses": {
-            "label": "Refresh license metadata",
-            "loading": "Refreshing license metadata for {typePlural}...",
-            "success": "Updated license metadata for {count} {typePlural}",
-            "none": "All {typePlural} already have license metadata",
-            "error": "Failed to refresh license metadata for {typePlural}: {message}"
+            "label": "刷新许可证元数据",
+            "loading": "正在刷新 {typePlural} 的许可证元数据...",
+            "success": "已更新 {count} 个 {typePlural} 的许可证元数据",
+            "none": "所有 {typePlural} 都已具备许可证元数据",
+            "error": "刷新 {typePlural} 的许可证元数据失败：{message}"
        },
        "repairRecipes": {
            "label": "修复配方数据",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "预设名称...",
            "baseModel": "基础模型",
            "modelTags": "标签（前20）",
-            "modelTypes": "Model Types",
+            "modelTypes": "模型类型",
            "license": "许可证",
            "noCreditRequired": "无需署名",
            "allowSellingGeneratedContent": "允许销售",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "最少"
            },
            "refresh": {
-                "title": "刷新配方列表"
+                "title": "刷新配方列表",
+                "quick": "同步变更",
+                "quickTooltip": "同步变更 - 快速刷新而不重建缓存",
+                "full": "重建缓存",
+                "fullTooltip": "重建缓存 - 重新扫描所有配方文件"
            },
            "filteredByLora": "按 LoRA 筛选",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "修复配方失败：{message}",
                "missingId": "无法修复配方：缺少配方 ID"
            }
+        },
+        "batchImport": {
+            "title": "批量导入配方",
+            "action": "批量导入",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "输入目录路径以导入该文件夹中的所有图片。",
+            "urlsLabel": "图片 URL 或本地路径",
+            "urlsPlaceholder": "https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "/图片/文件夹/路径",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "标签（可选，应用于所有配方）",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "跳过无元数据的图片",
+            "skipNoMetadataHelp": "没有 LoRA 元数据的图片将自动跳过。",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "开始导入",
+            "importing": "正在导入配方...",
+            "progress": "进度",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "当前",
+            "preparing": "准备中...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "取消",
+            "cancelled": "批量导入已取消",
+            "completed": "导入完成",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "成功导入 {count} 个配方",
+            "successCount": "成功",
+            "failedCount": "失败",
+            "skippedCount": "跳过",
+            "totalProcessed": "总计处理",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "请至少输入一个 URL 或路径",
+                "enterDirectory": "请输入目录路径",
+                "startFailed": "启动导入失败：{message}"
+            }
        }
    },
    "checkpoints": {
@@ -760,7 +823,7 @@
            "emptyFolderName": "请输入文件夹名称",
            "invalidFolderName": "文件夹名称包含无效字符",
            "noDragState": "未找到待处理的拖放操作"
-            },
+        },
        "empty": {
            "noFolders": "未找到文件夹",
            "dragHint": "拖拽项目到此处以创建文件夹"
@@ -1396,6 +1459,8 @@
            "loadFailed": "加载 {modelType} 失败：{message}",
            "refreshComplete": "刷新完成",
            "refreshFailed": "刷新配方失败：{message}",
+            "syncComplete": "同步完成",
+            "syncFailed": "同步配方失败：{message}",
            "updateFailed": "更新配方失败：{error}",
            "updateError": "更新配方出错：{message}",
            "nameSaved": "配方“{name}”保存成功",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "保存配方失败：{error}",
            "importFailed": "导入失败：{message}",
            "folderTreeFailed": "加载文件夹树失败",
-            "folderTreeError": "加载文件夹树出错"
+            "folderTreeError": "加载文件夹树出错",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "未选中模型",
--- a/locales/zh-TW.json
+++ b/locales/zh-TW.json
@@ -14,7 +14,8 @@
            "backToTop": "回到頂部",
            "settings": "設定",
            "help": "說明",
-            "add": "新增"
+            "add": "新增",
+            "close": "關閉"
        },
        "status": {
            "loading": "載入中...",
@@ -222,7 +223,7 @@
            "presetNamePlaceholder": "預設名稱...",
            "baseModel": "基礎模型",
            "modelTags": "標籤（前 20）",
-            "modelTypes": "Model Types",
+            "modelTypes": "模型類型",
            "license": "授權",
            "noCreditRequired": "無需署名",
            "allowSellingGeneratedContent": "允許銷售",
@@ -685,7 +686,11 @@
                "lorasCountAsc": "最少"
            },
            "refresh": {
-                "title": "重新整理配方列表"
+                "title": "重新整理配方列表",
+                "quick": "同步變更",
+                "quickTooltip": "同步變更 - 快速重新整理而不重建快取",
+                "full": "重建快取",
+                "fullTooltip": "重建快取 - 重新掃描所有配方檔案"
            },
            "filteredByLora": "已依 LoRA 篩選",
            "favorites": {
@@ -725,6 +730,64 @@
                "failed": "修復配方失敗：{message}",
                "missingId": "無法修復配方：缺少配方 ID"
            }
+        },
+        "batchImport": {
+            "title": "[TODO: Translate] Batch Import Recipes",
+            "action": "[TODO: Translate] Batch Import",
+            "urlList": "[TODO: Translate] URL List",
+            "directory": "[TODO: Translate] Directory",
+            "urlDescription": "[TODO: Translate] Enter image URLs or local file paths (one per line). Each will be imported as a recipe.",
+            "directoryDescription": "[TODO: Translate] Enter a directory path to import all images from that folder.",
+            "urlsLabel": "[TODO: Translate] Image URLs or Local Paths",
+            "urlsPlaceholder": "[TODO: Translate] https://civitai.com/images/...\nhttps://civitai.com/images/...\nC:/path/to/image.png\n...",
+            "urlsHint": "[TODO: Translate] Enter one URL or path per line",
+            "directoryPath": "[TODO: Translate] Directory Path",
+            "directoryPlaceholder": "[TODO: Translate] /path/to/images/folder",
+            "browse": "[TODO: Translate] Browse",
+            "recursive": "[TODO: Translate] Include subdirectories",
+            "tagsOptional": "[TODO: Translate] Tags (optional, applied to all recipes)",
+            "tagsPlaceholder": "[TODO: Translate] Enter tags separated by commas",
+            "tagsHint": "[TODO: Translate] Tags will be added to all imported recipes",
+            "skipNoMetadata": "[TODO: Translate] Skip images without metadata",
+            "skipNoMetadataHelp": "[TODO: Translate] Images without LoRA metadata will be skipped automatically.",
+            "start": "[TODO: Translate] Start Import",
+            "startImport": "[TODO: Translate] Start Import",
+            "importing": "[TODO: Translate] Importing...",
+            "progress": "[TODO: Translate] Progress",
+            "total": "[TODO: Translate] Total",
+            "success": "[TODO: Translate] Success",
+            "failed": "[TODO: Translate] Failed",
+            "skipped": "[TODO: Translate] Skipped",
+            "current": "[TODO: Translate] Current",
+            "currentItem": "[TODO: Translate] Current",
+            "preparing": "[TODO: Translate] Preparing...",
+            "cancel": "[TODO: Translate] Cancel",
+            "cancelImport": "[TODO: Translate] Cancel",
+            "cancelled": "[TODO: Translate] Import cancelled",
+            "completed": "[TODO: Translate] Import completed",
+            "completedWithErrors": "[TODO: Translate] Completed with errors",
+            "completedSuccess": "[TODO: Translate] Successfully imported {count} recipe(s)",
+            "successCount": "[TODO: Translate] Successful",
+            "failedCount": "[TODO: Translate] Failed",
+            "skippedCount": "[TODO: Translate] Skipped",
+            "totalProcessed": "[TODO: Translate] Total processed",
+            "viewDetails": "[TODO: Translate] View Details",
+            "newImport": "[TODO: Translate] New Import",
+            "manualPathEntry": "[TODO: Translate] Please enter the directory path manually. File browser is not available in this browser.",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {name}. You may need to enter the full path manually.",
+            "batchImportManualEntryRequired": "[TODO: Translate] File browser not available. Please enter the directory path manually.",
+            "backToParent": "[TODO: Translate] Back to parent directory",
+            "folders": "[TODO: Translate] Folders",
+            "folderCount": "[TODO: Translate] {count} folders",
+            "imageFiles": "[TODO: Translate] Image Files",
+            "images": "[TODO: Translate] images",
+            "imageCount": "[TODO: Translate] {count} images",
+            "selectFolder": "[TODO: Translate] Select This Folder",
+            "errors": {
+                "enterUrls": "[TODO: Translate] Please enter at least one URL or path",
+                "enterDirectory": "[TODO: Translate] Please enter a directory path",
+                "startFailed": "[TODO: Translate] Failed to start import: {message}"
+            }
        }
    },
    "checkpoints": {
@@ -1396,6 +1459,8 @@
            "loadFailed": "載入 {modelType} 失敗：{message}",
            "refreshComplete": "刷新完成",
            "refreshFailed": "刷新配方失敗：{message}",
+            "syncComplete": "同步完成",
+            "syncFailed": "同步配方失敗：{message}",
            "updateFailed": "更新配方失敗：{error}",
            "updateError": "更新配方錯誤：{message}",
            "nameSaved": "配方「{name}」已成功儲存",
@@ -1432,7 +1497,14 @@
            "recipeSaveFailed": "儲存配方失敗：{error}",
            "importFailed": "匯入失敗：{message}",
            "folderTreeFailed": "載入資料夾樹狀結構失敗",
-            "folderTreeError": "載入資料夾樹狀結構錯誤"
+            "folderTreeError": "載入資料夾樹狀結構錯誤",
+            "batchImportFailed": "[TODO: Translate] Failed to start batch import: {message}",
+            "batchImportCancelling": "[TODO: Translate] Cancelling batch import...",
+            "batchImportCancelFailed": "[TODO: Translate] Failed to cancel batch import: {message}",
+            "batchImportNoUrls": "[TODO: Translate] Please enter at least one URL or file path",
+            "batchImportNoDirectory": "[TODO: Translate] Please enter a directory path",
+            "batchImportBrowseFailed": "[TODO: Translate] Failed to browse directory: {message}",
+            "batchImportDirectorySelected": "[TODO: Translate] Directory selected: {path}"
        },
        "models": {
            "noModelsSelected": "未選擇模型",
--- a/py/config.py
+++ b/py/config.py
@@ -707,7 +707,13 @@ class Config:

    def _prepare_checkpoint_paths(
        self, checkpoint_paths: Iterable[str], unet_paths: Iterable[str]
-    ) -> List[str]:
+    ) -> Tuple[List[str], List[str], List[str]]:
+        """Prepare checkpoint paths and return (all_roots, checkpoint_roots, unet_roots).
+
+        Returns:
+            Tuple of (all_unique_paths, checkpoint_only_paths, unet_only_paths)
+            This method does NOT modify instance variables - callers must set them.
+        """
        checkpoint_map = self._dedupe_existing_paths(checkpoint_paths)
        unet_map = self._dedupe_existing_paths(unet_paths)

@@ -737,8 +743,8 @@ class Config:

        checkpoint_values = set(checkpoint_map.values())
        unet_values = set(unet_map.values())
-        self.checkpoints_roots = [p for p in unique_paths if p in checkpoint_values]
-        self.unet_roots = [p for p in unique_paths if p in unet_values]
+        checkpoint_roots = [p for p in unique_paths if p in checkpoint_values]
+        unet_roots = [p for p in unique_paths if p in unet_values]

        for original_path in unique_paths:
            real_path = os.path.normpath(os.path.realpath(original_path)).replace(
@@ -747,7 +753,7 @@ class Config:
            if real_path != original_path:
                self.add_path_mapping(original_path, real_path)

-        return unique_paths
+        return unique_paths, checkpoint_roots, unet_roots

    def _prepare_embedding_paths(self, raw_paths: Iterable[str]) -> List[str]:
        path_map = self._dedupe_existing_paths(raw_paths)
@@ -776,9 +782,11 @@ class Config:
        embedding_paths = folder_paths.get("embeddings", []) or []

        self.loras_roots = self._prepare_lora_paths(lora_paths)
-        self.base_models_roots = self._prepare_checkpoint_paths(
-            checkpoint_paths, unet_paths
-        )
+        (
+            self.base_models_roots,
+            self.checkpoints_roots,
+            self.unet_roots,
+        ) = self._prepare_checkpoint_paths(checkpoint_paths, unet_paths)
        self.embeddings_roots = self._prepare_embedding_paths(embedding_paths)

        # Process extra paths (only for LoRA Manager, not shared with ComfyUI)
@@ -789,18 +797,11 @@ class Config:
        extra_embedding_paths = extra_paths.get("embeddings", []) or []

        self.extra_loras_roots = self._prepare_lora_paths(extra_lora_paths)
-        # Save main paths before processing extra paths ( _prepare_checkpoint_paths overwrites them)
-        saved_checkpoints_roots = self.checkpoints_roots
-        saved_unet_roots = self.unet_roots
-        self.extra_checkpoints_roots = self._prepare_checkpoint_paths(
-            extra_checkpoint_paths, extra_unet_paths
-        )
-        self.extra_unet_roots = (
-            self.unet_roots if self.unet_roots is not None else []
-        )  # unet_roots was set by _prepare_checkpoint_paths
-        # Restore main paths
-        self.checkpoints_roots = saved_checkpoints_roots
-        self.unet_roots = saved_unet_roots
+        (
+            _,
+            self.extra_checkpoints_roots,
+            self.extra_unet_roots,
+        ) = self._prepare_checkpoint_paths(extra_checkpoint_paths, extra_unet_paths)
        self.extra_embeddings_roots = self._prepare_embedding_paths(
            extra_embedding_paths
        )
@@ -857,9 +858,11 @@ class Config:
        try:
            raw_checkpoint_paths = folder_paths.get_folder_paths("checkpoints")
            raw_unet_paths = folder_paths.get_folder_paths("unet")
-            unique_paths = self._prepare_checkpoint_paths(
-                raw_checkpoint_paths, raw_unet_paths
-            )
+            (
+                unique_paths,
+                self.checkpoints_roots,
+                self.unet_roots,
+            ) = self._prepare_checkpoint_paths(raw_checkpoint_paths, raw_unet_paths)

            logger.info(
                "Found checkpoint roots:"
--- a/py/nodes/checkpoint_loader.py
+++ b/py/nodes/checkpoint_loader.py
@@ -0,0 +1,184 @@
+import logging
+import os
+from typing import List, Tuple
+import comfy.sd
+import folder_paths
+from ..utils.utils import get_checkpoint_info_absolute, _format_model_name_for_comfyui
+
+logger = logging.getLogger(__name__)
+
+
+class CheckpointLoaderLM:
+    """Checkpoint Loader with support for extra folder paths
+
+    Loads checkpoints from both standard ComfyUI folders and LoRA Manager's
+    extra folder paths, providing a unified interface for checkpoint loading.
+    """
+
+    NAME = "CheckpointLoaderLM"
+    CATEGORY = "Lora Manager/loaders"
+
+    @classmethod
+    def INPUT_TYPES(s):
+        # Get list of checkpoint names from scanner (includes extra folder paths)
+        checkpoint_names = s._get_checkpoint_names()
+        return {
+            "required": {
+                "ckpt_name": (
+                    checkpoint_names,
+                    {"tooltip": "The name of the checkpoint (model) to load."},
+                ),
+            }
+        }
+
+    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
+    RETURN_NAMES = ("MODEL", "CLIP", "VAE")
+    OUTPUT_TOOLTIPS = (
+        "The model used for denoising latents.",
+        "The CLIP model used for encoding text prompts.",
+        "The VAE model used for encoding and decoding images to and from latent space.",
+    )
+    FUNCTION = "load_checkpoint"
+
+    @classmethod
+    def _get_checkpoint_names(cls) -> List[str]:
+        """Get list of checkpoint names from scanner cache in ComfyUI format (relative path with extension)"""
+        try:
+            from ..services.service_registry import ServiceRegistry
+            import asyncio
+
+            async def _get_names():
+                scanner = await ServiceRegistry.get_checkpoint_scanner()
+                cache = await scanner.get_cached_data()
+
+                # Get all model roots for calculating relative paths
+                model_roots = scanner.get_model_roots()
+
+                # Filter only checkpoint type (not diffusion_model) and format names
+                names = []
+                for item in cache.raw_data:
+                    if item.get("sub_type") == "checkpoint":
+                        file_path = item.get("file_path", "")
+                        if file_path:
+                            # Format as ComfyUI-style: "folder/model_name.ext"
+                            formatted_name = _format_model_name_for_comfyui(
+                                file_path, model_roots
+                            )
+                            if formatted_name:
+                                names.append(formatted_name)
+
+                return sorted(names)
+
+            try:
+                loop = asyncio.get_running_loop()
+                import concurrent.futures
+
+                def run_in_thread():
+                    new_loop = asyncio.new_event_loop()
+                    asyncio.set_event_loop(new_loop)
+                    try:
+                        return new_loop.run_until_complete(_get_names())
+                    finally:
+                        new_loop.close()
+
+                with concurrent.futures.ThreadPoolExecutor() as executor:
+                    future = executor.submit(run_in_thread)
+                    return future.result()
+            except RuntimeError:
+                return asyncio.run(_get_names())
+        except Exception as e:
+            logger.error(f"Error getting checkpoint names: {e}")
+            return []
+
+    def load_checkpoint(self, ckpt_name: str) -> Tuple:
+        """Load a checkpoint by name, supporting extra folder paths
+
+        Args:
+            ckpt_name: The name of the checkpoint to load (format: "folder/model_name.ext")
+
+        Returns:
+            Tuple of (MODEL, CLIP, VAE)
+        """
+        # Get absolute path from cache using ComfyUI-style name
+        ckpt_path, metadata = get_checkpoint_info_absolute(ckpt_name)
+
+        if metadata is None:
+            raise FileNotFoundError(
+                f"Checkpoint '{ckpt_name}' not found in LoRA Manager cache. "
+                "Make sure the checkpoint is indexed and try again."
+            )
+
+        # Check if it's a GGUF model
+        if ckpt_path.endswith(".gguf"):
+            return self._load_gguf_checkpoint(ckpt_path, ckpt_name)
+
+        # Load regular checkpoint using ComfyUI's API
+        logger.info(f"Loading checkpoint from: {ckpt_path}")
+        out = comfy.sd.load_checkpoint_guess_config(
+            ckpt_path,
+            output_vae=True,
+            output_clip=True,
+            embedding_directory=folder_paths.get_folder_paths("embeddings"),
+        )
+        return out[:3]
+
+    def _load_gguf_checkpoint(self, ckpt_path: str, ckpt_name: str) -> Tuple:
+        """Load a GGUF format checkpoint
+
+        Args:
+            ckpt_path: Absolute path to the GGUF file
+            ckpt_name: Name of the checkpoint for error messages
+
+        Returns:
+            Tuple of (MODEL, CLIP, VAE) - CLIP and VAE may be None for GGUF
+        """
+        try:
+            # Try to import ComfyUI-GGUF modules
+            from custom_nodes.ComfyUI_GGUF.loader import gguf_sd_loader
+            from custom_nodes.ComfyUI_GGUF.ops import GGMLOps
+            from custom_nodes.ComfyUI_GGUF.nodes import GGUFModelPatcher
+        except ImportError:
+            raise RuntimeError(
+                f"Cannot load GGUF model '{ckpt_name}'. "
+                "ComfyUI-GGUF is not installed. "
+                "Please install ComfyUI-GGUF from https://github.com/city96/ComfyUI-GGUF "
+                "to load GGUF format models."
+            )
+
+        logger.info(f"Loading GGUF checkpoint from: {ckpt_path}")
+
+        try:
+            # Load GGUF state dict
+            sd, extra = gguf_sd_loader(ckpt_path)
+
+            # Prepare kwargs for metadata if supported
+            kwargs = {}
+            import inspect
+
+            valid_params = inspect.signature(
+                comfy.sd.load_diffusion_model_state_dict
+            ).parameters
+            if "metadata" in valid_params:
+                kwargs["metadata"] = extra.get("metadata", {})
+
+            # Load the model
+            model = comfy.sd.load_diffusion_model_state_dict(
+                sd, model_options={"custom_operations": GGMLOps()}, **kwargs
+            )
+
+            if model is None:
+                raise RuntimeError(
+                    f"Could not detect model type for GGUF checkpoint: {ckpt_path}"
+                )
+
+            # Wrap with GGUFModelPatcher
+            model = GGUFModelPatcher.clone(model)
+
+            # GGUF checkpoints typically don't include CLIP/VAE
+            return (model, None, None)
+
+        except Exception as e:
+            logger.error(f"Error loading GGUF checkpoint '{ckpt_name}': {e}")
+            raise RuntimeError(
+                f"Failed to load GGUF checkpoint '{ckpt_name}': {str(e)}"
+            )
--- a/py/nodes/unet_loader.py
+++ b/py/nodes/unet_loader.py
@@ -0,0 +1,205 @@
+import logging
+import os
+from typing import List, Tuple
+import torch
+import comfy.sd
+from ..utils.utils import get_checkpoint_info_absolute, _format_model_name_for_comfyui
+
+logger = logging.getLogger(__name__)
+
+
+class UNETLoaderLM:
+    """UNET Loader with support for extra folder paths
+
+    Loads diffusion models/UNets from both standard ComfyUI folders and LoRA Manager's
+    extra folder paths, providing a unified interface for UNET loading.
+    Supports both regular diffusion models and GGUF format models.
+    """
+
+    NAME = "UNETLoaderLM"
+    CATEGORY = "Lora Manager/loaders"
+
+    @classmethod
+    def INPUT_TYPES(s):
+        # Get list of unet names from scanner (includes extra folder paths)
+        unet_names = s._get_unet_names()
+        return {
+            "required": {
+                "unet_name": (
+                    unet_names,
+                    {"tooltip": "The name of the diffusion model to load."},
+                ),
+                "weight_dtype": (
+                    ["default", "fp8_e4m3fn", "fp8_e4m3fn_fast", "fp8_e5m2"],
+                    {"tooltip": "The dtype to use for the model weights."},
+                ),
+            }
+        }
+
+    RETURN_TYPES = ("MODEL",)
+    RETURN_NAMES = ("MODEL",)
+    OUTPUT_TOOLTIPS = ("The model used for denoising latents.",)
+    FUNCTION = "load_unet"
+
+    @classmethod
+    def _get_unet_names(cls) -> List[str]:
+        """Get list of diffusion model names from scanner cache in ComfyUI format (relative path with extension)"""
+        try:
+            from ..services.service_registry import ServiceRegistry
+            import asyncio
+
+            async def _get_names():
+                scanner = await ServiceRegistry.get_checkpoint_scanner()
+                cache = await scanner.get_cached_data()
+
+                # Get all model roots for calculating relative paths
+                model_roots = scanner.get_model_roots()
+
+                # Filter only diffusion_model type and format names
+                names = []
+                for item in cache.raw_data:
+                    if item.get("sub_type") == "diffusion_model":
+                        file_path = item.get("file_path", "")
+                        if file_path:
+                            # Format as ComfyUI-style: "folder/model_name.ext"
+                            formatted_name = _format_model_name_for_comfyui(
+                                file_path, model_roots
+                            )
+                            if formatted_name:
+                                names.append(formatted_name)
+
+                return sorted(names)
+
+            try:
+                loop = asyncio.get_running_loop()
+                import concurrent.futures
+
+                def run_in_thread():
+                    new_loop = asyncio.new_event_loop()
+                    asyncio.set_event_loop(new_loop)
+                    try:
+                        return new_loop.run_until_complete(_get_names())
+                    finally:
+                        new_loop.close()
+
+                with concurrent.futures.ThreadPoolExecutor() as executor:
+                    future = executor.submit(run_in_thread)
+                    return future.result()
+            except RuntimeError:
+                return asyncio.run(_get_names())
+        except Exception as e:
+            logger.error(f"Error getting unet names: {e}")
+            return []
+
+    def load_unet(self, unet_name: str, weight_dtype: str) -> Tuple:
+        """Load a diffusion model by name, supporting extra folder paths
+
+        Args:
+            unet_name: The name of the diffusion model to load (format: "folder/model_name.ext")
+            weight_dtype: The dtype to use for model weights
+
+        Returns:
+            Tuple of (MODEL,)
+        """
+        # Get absolute path from cache using ComfyUI-style name
+        unet_path, metadata = get_checkpoint_info_absolute(unet_name)
+
+        if metadata is None:
+            raise FileNotFoundError(
+                f"Diffusion model '{unet_name}' not found in LoRA Manager cache. "
+                "Make sure the model is indexed and try again."
+            )
+
+        # Check if it's a GGUF model
+        if unet_path.endswith(".gguf"):
+            return self._load_gguf_unet(unet_path, unet_name, weight_dtype)
+
+        # Load regular diffusion model using ComfyUI's API
+        logger.info(f"Loading diffusion model from: {unet_path}")
+
+        # Build model options based on weight_dtype
+        model_options = {}
+        if weight_dtype == "fp8_e4m3fn":
+            model_options["dtype"] = torch.float8_e4m3fn
+        elif weight_dtype == "fp8_e4m3fn_fast":
+            model_options["dtype"] = torch.float8_e4m3fn
+            model_options["fp8_optimizations"] = True
+        elif weight_dtype == "fp8_e5m2":
+            model_options["dtype"] = torch.float8_e5m2
+
+        model = comfy.sd.load_diffusion_model(unet_path, model_options=model_options)
+        return (model,)
+
+    def _load_gguf_unet(
+        self, unet_path: str, unet_name: str, weight_dtype: str
+    ) -> Tuple:
+        """Load a GGUF format diffusion model
+
+        Args:
+            unet_path: Absolute path to the GGUF file
+            unet_name: Name of the model for error messages
+            weight_dtype: The dtype to use for model weights
+
+        Returns:
+            Tuple of (MODEL,)
+        """
+        try:
+            # Try to import ComfyUI-GGUF modules
+            from custom_nodes.ComfyUI_GGUF.loader import gguf_sd_loader
+            from custom_nodes.ComfyUI_GGUF.ops import GGMLOps
+            from custom_nodes.ComfyUI_GGUF.nodes import GGUFModelPatcher
+        except ImportError:
+            raise RuntimeError(
+                f"Cannot load GGUF model '{unet_name}'. "
+                "ComfyUI-GGUF is not installed. "
+                "Please install ComfyUI-GGUF from https://github.com/city96/ComfyUI-GGUF "
+                "to load GGUF format models."
+            )
+
+        logger.info(f"Loading GGUF diffusion model from: {unet_path}")
+
+        try:
+            # Load GGUF state dict
+            sd, extra = gguf_sd_loader(unet_path)
+
+            # Prepare kwargs for metadata if supported
+            kwargs = {}
+            import inspect
+
+            valid_params = inspect.signature(
+                comfy.sd.load_diffusion_model_state_dict
+            ).parameters
+            if "metadata" in valid_params:
+                kwargs["metadata"] = extra.get("metadata", {})
+
+            # Setup custom operations with GGUF support
+            ops = GGMLOps()
+
+            # Handle weight_dtype for GGUF models
+            if weight_dtype in ("default", None):
+                ops.Linear.dequant_dtype = None
+            elif weight_dtype in ["target"]:
+                ops.Linear.dequant_dtype = weight_dtype
+            else:
+                ops.Linear.dequant_dtype = getattr(torch, weight_dtype, None)
+
+            # Load the model
+            model = comfy.sd.load_diffusion_model_state_dict(
+                sd, model_options={"custom_operations": ops}, **kwargs
+            )
+
+            if model is None:
+                raise RuntimeError(
+                    f"Could not detect model type for GGUF diffusion model: {unet_path}"
+                )
+
+            # Wrap with GGUFModelPatcher
+            model = GGUFModelPatcher.clone(model)
+
+            return (model,)
+
+        except Exception as e:
+            logger.error(f"Error loading GGUF diffusion model '{unet_name}': {e}")
+            raise RuntimeError(
+                f"Failed to load GGUF diffusion model '{unet_name}': {str(e)}"
+            )
--- a/py/routes/base_recipe_routes.py
+++ b/py/routes/base_recipe_routes.py
@@ -1,4 +1,5 @@
 """Base infrastructure shared across recipe routes."""
+
 from __future__ import annotations

 import logging
@@ -16,12 +17,14 @@ from ..services.recipes import (
    RecipePersistenceService,
    RecipeSharingService,
 )
+from ..services.batch_import_service import BatchImportService
 from ..services.server_i18n import server_i18n
 from ..services.service_registry import ServiceRegistry
 from ..services.settings_manager import get_settings_manager
 from ..utils.constants import CARD_PREVIEW_WIDTH
 from ..utils.exif_utils import ExifUtils
 from .handlers.recipe_handlers import (
+    BatchImportHandler,
    RecipeAnalysisHandler,
    RecipeHandlerSet,
    RecipeListingHandler,
@@ -116,7 +119,10 @@ class BaseRecipeRoutes:
        recipe_scanner_getter = lambda: self.recipe_scanner
        civitai_client_getter = lambda: self.civitai_client

-        standalone_mode = os.environ.get("LORA_MANAGER_STANDALONE", "0") == "1" or os.environ.get("HF_HUB_DISABLE_TELEMETRY", "0") == "0"
+        standalone_mode = (
+            os.environ.get("LORA_MANAGER_STANDALONE", "0") == "1"
+            or os.environ.get("HF_HUB_DISABLE_TELEMETRY", "0") == "0"
+        )
        if not standalone_mode:
            from ..metadata_collector import get_metadata  # type: ignore[import-not-found]
            from ..metadata_collector.metadata_processor import (  # type: ignore[import-not-found]
@@ -190,6 +196,22 @@ class BaseRecipeRoutes:
            sharing_service=sharing_service,
        )

+        from ..services.websocket_manager import ws_manager
+
+        batch_import_service = BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+        batch_import = BatchImportHandler(
+            ensure_dependencies_ready=self.ensure_dependencies_ready,
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            logger=logger,
+            batch_import_service=batch_import_service,
+        )
+
        return RecipeHandlerSet(
            page_view=page_view,
            listing=listing,
@@ -197,4 +219,5 @@ class BaseRecipeRoutes:
            management=management,
            analysis=analysis,
            sharing=sharing,
+            batch_import=batch_import,
        )
--- a/py/routes/handlers/misc_handlers.py
+++ b/py/routes/handlers/misc_handlers.py
@@ -240,11 +240,7 @@ class SupportersHandler:
        except Exception as e:
            self._logger.debug(f"Failed to load supporters data: {e}")

-        return {
-            "specialThanks": [],
-            "allSupporters": [],
-            "totalCount": 0
-        }
+        return {"specialThanks": [], "allSupporters": [], "totalCount": 0}

    async def get_supporters(self, request: web.Request) -> web.Response:
        """Return supporters data as JSON."""
@@ -253,9 +249,101 @@ class SupportersHandler:
            return web.json_response({"success": True, "supporters": supporters})
        except Exception as exc:
            self._logger.error("Error loading supporters: %s", exc, exc_info=True)
-            return web.json_response(
-                {"success": False, "error": str(exc)}, status=500
+            return web.json_response({"success": False, "error": str(exc)}, status=500)
+
+
+class ExampleWorkflowsHandler:
+    """Handler for example workflow templates."""
+
+    def __init__(self, logger: logging.Logger | None = None) -> None:
+        self._logger = logger or logging.getLogger(__name__)
+
+    def _get_workflows_dir(self) -> str:
+        """Get the example workflows directory path."""
+        current_file = os.path.abspath(__file__)
+        root_dir = os.path.dirname(
+            os.path.dirname(os.path.dirname(os.path.dirname(current_file)))
+        )
+        return os.path.join(root_dir, "example_workflows")
+
+    def _format_workflow_name(self, filename: str) -> str:
+        """Convert filename to human-readable name."""
+        name = os.path.splitext(filename)[0]
+        name = name.replace("_", " ")
+        return name
+
+    async def get_example_workflows(self, request: web.Request) -> web.Response:
+        """Return list of available example workflows."""
+        try:
+            workflows_dir = self._get_workflows_dir()
+            workflows = [
+                {
+                    "value": "Default",
+                    "label": "Default (Blank)",
+                    "path": None,
+                }
+            ]
+
+            if os.path.exists(workflows_dir):
+                for filename in sorted(os.listdir(workflows_dir)):
+                    if filename.endswith(".json"):
+                        workflows.append(
+                            {
+                                "value": filename,
+                                "label": self._format_workflow_name(filename),
+                                "path": f"example_workflows/{filename}",
+                            }
+                        )
+
+            return web.json_response({"success": True, "workflows": workflows})
+        except Exception as exc:
+            self._logger.error(
+                "Error listing example workflows: %s", exc, exc_info=True
            )
+            return web.json_response({"success": False, "error": str(exc)}, status=500)
+
+    async def get_example_workflow(self, request: web.Request) -> web.Response:
+        """Return a specific example workflow JSON content."""
+        try:
+            filename = request.match_info.get("filename")
+            if not filename:
+                return web.json_response(
+                    {"success": False, "error": "Filename not provided"}, status=400
+                )
+
+            if filename == "Default":
+                return web.json_response(
+                    {
+                        "success": True,
+                        "workflow": {
+                            "last_node_id": 0,
+                            "last_link_id": 0,
+                            "nodes": [],
+                            "links": [],
+                            "groups": [],
+                            "config": {},
+                            "extra": {},
+                            "version": 0.4,
+                        },
+                    }
+                )
+
+            workflows_dir = self._get_workflows_dir()
+            filepath = os.path.join(workflows_dir, filename)
+
+            if not os.path.exists(filepath):
+                return web.json_response(
+                    {"success": False, "error": f"Workflow not found: {filename}"},
+                    status=404,
+                )
+
+            with open(filepath, "r", encoding="utf-8") as f:
+                workflow = json.load(f)
+
+            return web.json_response({"success": True, "workflow": workflow})
+        except Exception as exc:
+            self._logger.error("Error loading example workflow: %s", exc, exc_info=True)
+            return web.json_response({"success": False, "error": str(exc)}, status=500)


 class SettingsHandler:
@@ -263,15 +351,17 @@ class SettingsHandler:

    # Settings keys that should NOT be synced to frontend.
    # All other settings are synced by default.
-    _NO_SYNC_KEYS = frozenset({
-        # Internal/performance settings (not used by frontend)
-        "hash_chunk_size_mb",
-        "download_stall_timeout_seconds",
-        # Complex internal structures retrieved via separate endpoints
-        "folder_paths",
-        "libraries",
-        "active_library",
-    })
+    _NO_SYNC_KEYS = frozenset(
+        {
+            # Internal/performance settings (not used by frontend)
+            "hash_chunk_size_mb",
+            "download_stall_timeout_seconds",
+            # Complex internal structures retrieved via separate endpoints
+            "folder_paths",
+            "libraries",
+            "active_library",
+        }
+    )

    _PROXY_KEYS = {
        "proxy_enabled",
@@ -1226,6 +1316,7 @@ class CustomWordsHandler:

    def __init__(self) -> None:
        from ...services.custom_words_service import get_custom_words_service
+
        self._service = get_custom_words_service()

    async def search_custom_words(self, request: web.Request) -> web.Response:
@@ -1234,6 +1325,7 @@ class CustomWordsHandler:
        Query parameters:
            search: The search term to match against.
            limit: Maximum number of results to return (default: 20).
+            offset: Number of results to skip (default: 0).
            category: Optional category filter. Can be:
                - A category name (e.g., "character", "artist", "general")
                - Comma-separated category IDs (e.g., "4,11" for character)
@@ -1243,6 +1335,7 @@ class CustomWordsHandler:
        try:
            search_term = request.query.get("search", "")
            limit = int(request.query.get("limit", "20"))
+            offset = max(0, int(request.query.get("offset", "0")))
            category_param = request.query.get("category", "")
            enriched_param = request.query.get("enriched", "").lower() == "true"

@@ -1252,13 +1345,14 @@ class CustomWordsHandler:
                categories = self._parse_category_param(category_param)

            results = self._service.search_words(
-                search_term, limit, categories=categories, enriched=enriched_param
+                search_term,
+                limit,
+                offset=offset,
+                categories=categories,
+                enriched=enriched_param,
            )

-            return web.json_response({
-                "success": True,
-                "words": results
-            })
+            return web.json_response({"success": True, "words": results})
        except Exception as exc:
            logger.error("Error searching custom words: %s", exc, exc_info=True)
            return web.json_response({"error": str(exc)}, status=500)
@@ -1523,6 +1617,7 @@ class MiscHandlerSet:
        filesystem: FileSystemHandler,
        custom_words: CustomWordsHandler,
        supporters: SupportersHandler,
+        example_workflows: ExampleWorkflowsHandler,
    ) -> None:
        self.health = health
        self.settings = settings
@@ -1536,6 +1631,7 @@ class MiscHandlerSet:
        self.filesystem = filesystem
        self.custom_words = custom_words
        self.supporters = supporters
+        self.example_workflows = example_workflows

    def to_route_mapping(
        self,
@@ -1565,6 +1661,8 @@ class MiscHandlerSet:
            "open_settings_location": self.filesystem.open_settings_location,
            "search_custom_words": self.custom_words.search_custom_words,
            "get_supporters": self.supporters.get_supporters,
+            "get_example_workflows": self.example_workflows.get_example_workflows,
+            "get_example_workflow": self.example_workflows.get_example_workflow,
        }


--- a/py/routes/handlers/model_handlers.py
+++ b/py/routes/handlers/model_handlers.py
@@ -1268,8 +1268,11 @@ class ModelQueryHandler:
    async def get_relative_paths(self, request: web.Request) -> web.Response:
        try:
            search = request.query.get("search", "").strip()
-            limit = min(int(request.query.get("limit", "15")), 50)
-            matching_paths = await self._service.search_relative_paths(search, limit)
+            limit = min(int(request.query.get("limit", "15")), 100)
+            offset = max(0, int(request.query.get("offset", "0")))
+            matching_paths = await self._service.search_relative_paths(
+                search, limit, offset
+            )
            return web.json_response(
                {"success": True, "relative_paths": matching_paths}
            )
--- a/py/routes/handlers/recipe_handlers.py
+++ b/py/routes/handlers/recipe_handlers.py
@@ -1,4 +1,5 @@
 """Dedicated handler objects for recipe-related routes."""
+
 from __future__ import annotations

 import json
@@ -8,6 +9,7 @@ import re
 import asyncio
 import tempfile
 from dataclasses import dataclass
+from pathlib import Path
 from typing import Any, Awaitable, Callable, Dict, List, Mapping, Optional

 from aiohttp import web
@@ -29,6 +31,7 @@ from ...utils.exif_utils import ExifUtils
 from ...recipes.merger import GenParamsMerger
 from ...recipes.enrichment import RecipeEnricher
 from ...services.websocket_manager import ws_manager as default_ws_manager
+from ...services.batch_import_service import BatchImportService

 Logger = logging.Logger
 EnsureDependenciesCallable = Callable[[], Awaitable[None]]
@@ -46,8 +49,11 @@ class RecipeHandlerSet:
    management: "RecipeManagementHandler"
    analysis: "RecipeAnalysisHandler"
    sharing: "RecipeSharingHandler"
+    batch_import: "BatchImportHandler"

-    def to_route_mapping(self) -> Mapping[str, Callable[[web.Request], Awaitable[web.StreamResponse]]]:
+    def to_route_mapping(
+        self,
+    ) -> Mapping[str, Callable[[web.Request], Awaitable[web.StreamResponse]]]:
        """Expose handler coroutines keyed by registrar handler names."""

        return {
@@ -81,6 +87,11 @@ class RecipeHandlerSet:
            "cancel_repair": self.management.cancel_repair,
            "repair_recipe": self.management.repair_recipe,
            "get_repair_progress": self.management.get_repair_progress,
+            "start_batch_import": self.batch_import.start_batch_import,
+            "get_batch_import_progress": self.batch_import.get_batch_import_progress,
+            "cancel_batch_import": self.batch_import.cancel_batch_import,
+            "start_directory_import": self.batch_import.start_directory_import,
+            "browse_directory": self.batch_import.browse_directory,
        }


@@ -170,8 +181,10 @@ class RecipeListingHandler:
            search_options = {
                "title": request.query.get("search_title", "true").lower() == "true",
                "tags": request.query.get("search_tags", "true").lower() == "true",
-                "lora_name": request.query.get("search_lora_name", "true").lower() == "true",
-                "lora_model": request.query.get("search_lora_model", "true").lower() == "true",
+                "lora_name": request.query.get("search_lora_name", "true").lower()
+                == "true",
+                "lora_model": request.query.get("search_lora_model", "true").lower()
+                == "true",
                "prompt": request.query.get("search_prompt", "true").lower() == "true",
            }

@@ -246,7 +259,9 @@ class RecipeListingHandler:
                return web.json_response({"error": "Recipe not found"}, status=404)
            return web.json_response(recipe)
        except Exception as exc:
-            self._logger.error("Error retrieving recipe details: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error retrieving recipe details: %s", exc, exc_info=True
+            )
            return web.json_response({"error": str(exc)}, status=500)

    def format_recipe_file_url(self, file_path: str) -> str:
@@ -256,7 +271,9 @@ class RecipeListingHandler:
            if static_url:
                return static_url
        except Exception as exc:  # pragma: no cover - logging path
-            self._logger.error("Error formatting recipe file URL: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error formatting recipe file URL: %s", exc, exc_info=True
+            )
            return "/loras_static/images/no-preview.png"

        return "/loras_static/images/no-preview.png"
@@ -293,7 +310,9 @@ class RecipeQueryHandler:
                for tag in recipe.get("tags", []) or []:
                    tag_counts[tag] = tag_counts.get(tag, 0) + 1

-            sorted_tags = [{"tag": tag, "count": count} for tag, count in tag_counts.items()]
+            sorted_tags = [
+                {"tag": tag, "count": count} for tag, count in tag_counts.items()
+            ]
            sorted_tags.sort(key=lambda entry: entry["count"], reverse=True)
            return web.json_response({"success": True, "tags": sorted_tags[:limit]})
        except Exception as exc:
@@ -313,9 +332,14 @@ class RecipeQueryHandler:
            for recipe in getattr(cache, "raw_data", []):
                base_model = recipe.get("base_model")
                if base_model:
-                    base_model_counts[base_model] = base_model_counts.get(base_model, 0) + 1
+                    base_model_counts[base_model] = (
+                        base_model_counts.get(base_model, 0) + 1
+                    )

-            sorted_models = [{"name": model, "count": count} for model, count in base_model_counts.items()]
+            sorted_models = [
+                {"name": model, "count": count}
+                for model, count in base_model_counts.items()
+            ]
            sorted_models.sort(key=lambda entry: entry["count"], reverse=True)
            return web.json_response({"success": True, "base_models": sorted_models})
        except Exception as exc:
@@ -345,7 +369,9 @@ class RecipeQueryHandler:
            folders = await recipe_scanner.get_folders()
            return web.json_response({"success": True, "folders": folders})
        except Exception as exc:
-            self._logger.error("Error retrieving recipe folders: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error retrieving recipe folders: %s", exc, exc_info=True
+            )
            return web.json_response({"success": False, "error": str(exc)}, status=500)

    async def get_folder_tree(self, request: web.Request) -> web.Response:
@@ -358,7 +384,9 @@ class RecipeQueryHandler:
            folder_tree = await recipe_scanner.get_folder_tree()
            return web.json_response({"success": True, "tree": folder_tree})
        except Exception as exc:
-            self._logger.error("Error retrieving recipe folder tree: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error retrieving recipe folder tree: %s", exc, exc_info=True
+            )
            return web.json_response({"success": False, "error": str(exc)}, status=500)

    async def get_unified_folder_tree(self, request: web.Request) -> web.Response:
@@ -371,7 +399,9 @@ class RecipeQueryHandler:
            folder_tree = await recipe_scanner.get_folder_tree()
            return web.json_response({"success": True, "tree": folder_tree})
        except Exception as exc:
-            self._logger.error("Error retrieving unified recipe folder tree: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error retrieving unified recipe folder tree: %s", exc, exc_info=True
+            )
            return web.json_response({"success": False, "error": str(exc)}, status=500)

    async def get_recipes_for_lora(self, request: web.Request) -> web.Response:
@@ -383,7 +413,9 @@ class RecipeQueryHandler:

            lora_hash = request.query.get("hash")
            if not lora_hash:
-                return web.json_response({"success": False, "error": "Lora hash is required"}, status=400)
+                return web.json_response(
+                    {"success": False, "error": "Lora hash is required"}, status=400
+                )

            matching_recipes = await recipe_scanner.get_recipes_for_lora(lora_hash)
            return web.json_response({"success": True, "recipes": matching_recipes})
@@ -400,7 +432,9 @@ class RecipeQueryHandler:

            self._logger.info("Manually triggering recipe cache rebuild")
            await recipe_scanner.get_cached_data(force_refresh=True)
-            return web.json_response({"success": True, "message": "Recipe cache refreshed successfully"})
+            return web.json_response(
+                {"success": True, "message": "Recipe cache refreshed successfully"}
+            )
        except Exception as exc:
            self._logger.error("Error refreshing recipe cache: %s", exc, exc_info=True)
            return web.json_response({"success": False, "error": str(exc)}, status=500)
@@ -429,7 +463,9 @@ class RecipeQueryHandler:
                                "id": recipe.get("id"),
                                "title": recipe.get("title"),
                                "file_url": recipe.get("file_url")
-                                or self._format_recipe_file_url(recipe.get("file_path", "")),
+                                or self._format_recipe_file_url(
+                                    recipe.get("file_path", "")
+                                ),
                                "modified": recipe.get("modified"),
                                "created_date": recipe.get("created_date"),
                                "lora_count": len(recipe.get("loras", [])),
@@ -437,7 +473,9 @@ class RecipeQueryHandler:
                        )

                if len(recipes) >= 2:
-                    recipes.sort(key=lambda entry: entry.get("modified", 0), reverse=True)
+                    recipes.sort(
+                        key=lambda entry: entry.get("modified", 0), reverse=True
+                    )
                    response_data.append(
                        {
                            "type": "fingerprint",
@@ -460,7 +498,9 @@ class RecipeQueryHandler:
                                "id": recipe.get("id"),
                                "title": recipe.get("title"),
                                "file_url": recipe.get("file_url")
-                                or self._format_recipe_file_url(recipe.get("file_path", "")),
+                                or self._format_recipe_file_url(
+                                    recipe.get("file_path", "")
+                                ),
                                "modified": recipe.get("modified"),
                                "created_date": recipe.get("created_date"),
                                "lora_count": len(recipe.get("loras", [])),
@@ -468,7 +508,9 @@ class RecipeQueryHandler:
                        )

                if len(recipes) >= 2:
-                    recipes.sort(key=lambda entry: entry.get("modified", 0), reverse=True)
+                    recipes.sort(
+                        key=lambda entry: entry.get("modified", 0), reverse=True
+                    )
                    response_data.append(
                        {
                            "type": "source_url",
@@ -479,9 +521,13 @@ class RecipeQueryHandler:
                    )

            response_data.sort(key=lambda entry: entry["count"], reverse=True)
-            return web.json_response({"success": True, "duplicate_groups": response_data})
+            return web.json_response(
+                {"success": True, "duplicate_groups": response_data}
+            )
        except Exception as exc:
-            self._logger.error("Error finding duplicate recipes: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error finding duplicate recipes: %s", exc, exc_info=True
+            )
            return web.json_response({"success": False, "error": str(exc)}, status=500)

    async def get_recipe_syntax(self, request: web.Request) -> web.Response:
@@ -498,9 +544,13 @@ class RecipeQueryHandler:
                return web.json_response({"error": "Recipe not found"}, status=404)

            if not syntax_parts:
-                return web.json_response({"error": "No LoRAs found in this recipe"}, status=400)
+                return web.json_response(
+                    {"error": "No LoRAs found in this recipe"}, status=400
+                )

-            return web.json_response({"success": True, "syntax": " ".join(syntax_parts)})
+            return web.json_response(
+                {"success": True, "syntax": " ".join(syntax_parts)}
+            )
        except Exception as exc:
            self._logger.error("Error generating recipe syntax: %s", exc, exc_info=True)
            return web.json_response({"error": str(exc)}, status=500)
@@ -561,11 +611,17 @@ class RecipeManagementHandler:
            await self._ensure_dependencies_ready()
            recipe_scanner = self._recipe_scanner_getter()
            if recipe_scanner is None:
-                return web.json_response({"success": False, "error": "Recipe scanner unavailable"}, status=503)
+                return web.json_response(
+                    {"success": False, "error": "Recipe scanner unavailable"},
+                    status=503,
+                )

            # Check if already running
            if self._ws_manager.is_recipe_repair_running():
-                return web.json_response({"success": False, "error": "Recipe repair already in progress"}, status=409)
+                return web.json_response(
+                    {"success": False, "error": "Recipe repair already in progress"},
+                    status=409,
+                )

            recipe_scanner.reset_cancellation()

@@ -579,11 +635,12 @@ class RecipeManagementHandler:
                        progress_callback=progress_callback
                    )
                except Exception as e:
-                    self._logger.error(f"Error in recipe repair task: {e}", exc_info=True)
-                    await self._ws_manager.broadcast_recipe_repair_progress({
-                        "status": "error",
-                        "error": str(e)
-                    })
+                    self._logger.error(
+                        f"Error in recipe repair task: {e}", exc_info=True
+                    )
+                    await self._ws_manager.broadcast_recipe_repair_progress(
+                        {"status": "error", "error": str(e)}
+                    )
                finally:
                    # Keep the final status for a while so the UI can see it
                    await asyncio.sleep(5)
@@ -593,7 +650,9 @@ class RecipeManagementHandler:

            asyncio.create_task(run_repair())

-            return web.json_response({"success": True, "message": "Recipe repair started"})
+            return web.json_response(
+                {"success": True, "message": "Recipe repair started"}
+            )
        except Exception as exc:
            self._logger.error("Error starting recipe repair: %s", exc, exc_info=True)
            return web.json_response({"success": False, "error": str(exc)}, status=500)
@@ -603,10 +662,15 @@ class RecipeManagementHandler:
            await self._ensure_dependencies_ready()
            recipe_scanner = self._recipe_scanner_getter()
            if recipe_scanner is None:
-                return web.json_response({"success": False, "error": "Recipe scanner unavailable"}, status=503)
+                return web.json_response(
+                    {"success": False, "error": "Recipe scanner unavailable"},
+                    status=503,
+                )

            recipe_scanner.cancel_task()
-            return web.json_response({"success": True, "message": "Cancellation requested"})
+            return web.json_response(
+                {"success": True, "message": "Cancellation requested"}
+            )
        except Exception as exc:
            self._logger.error("Error cancelling recipe repair: %s", exc, exc_info=True)
            return web.json_response({"success": False, "error": str(exc)}, status=500)
@@ -616,7 +680,10 @@ class RecipeManagementHandler:
            await self._ensure_dependencies_ready()
            recipe_scanner = self._recipe_scanner_getter()
            if recipe_scanner is None:
-                return web.json_response({"success": False, "error": "Recipe scanner unavailable"}, status=503)
+                return web.json_response(
+                    {"success": False, "error": "Recipe scanner unavailable"},
+                    status=503,
+                )

            recipe_id = request.match_info["recipe_id"]
            result = await recipe_scanner.repair_recipe_by_id(recipe_id)
@@ -632,25 +699,26 @@ class RecipeManagementHandler:
            progress = self._ws_manager.get_recipe_repair_progress()
            if progress:
                return web.json_response({"success": True, "progress": progress})
-            return web.json_response({"success": False, "message": "No repair in progress"}, status=404)
+            return web.json_response(
+                {"success": False, "message": "No repair in progress"}, status=404
+            )
        except Exception as exc:
            self._logger.error("Error getting repair progress: %s", exc, exc_info=True)
            return web.json_response({"success": False, "error": str(exc)}, status=500)

-
    async def import_remote_recipe(self, request: web.Request) -> web.Response:
        try:
            await self._ensure_dependencies_ready()
            recipe_scanner = self._recipe_scanner_getter()
            if recipe_scanner is None:
                raise RuntimeError("Recipe scanner unavailable")
-            
+
            # 1. Parse Parameters
            params = request.rel_url.query
            image_url = params.get("image_url")
            name = params.get("name")
            resources_raw = params.get("resources")
-            
+
            if not image_url:
                raise RecipeValidationError("Missing required field: image_url")
            if not name:
@@ -658,64 +726,80 @@ class RecipeManagementHandler:
            if not resources_raw:
                raise RecipeValidationError("Missing required field: resources")

-            checkpoint_entry, lora_entries = self._parse_resources_payload(resources_raw)
+            checkpoint_entry, lora_entries = self._parse_resources_payload(
+                resources_raw
+            )
            gen_params_request = self._parse_gen_params(params.get("gen_params"))
-            
+
            # 2. Initial Metadata Construction
            metadata: Dict[str, Any] = {
                "base_model": params.get("base_model", "") or "",
                "loras": lora_entries,
                "gen_params": gen_params_request or {},
-                "source_url": image_url
+                "source_url": image_url,
            }
-            
+
            source_path = params.get("source_path")
            if source_path:
                metadata["source_path"] = source_path
-                
+
            # Checkpoint handling
            if checkpoint_entry:
                metadata["checkpoint"] = checkpoint_entry
                # Ensure checkpoint is also in gen_params for consistency if needed by enricher?
                # Actually enricher looks at metadata['checkpoint'], so this is fine.
-                
+
                # Try to resolve base model from checkpoint if not explicitly provided
                if not metadata["base_model"]:
-                    base_model_from_metadata = await self._resolve_base_model_from_checkpoint(checkpoint_entry)
+                    base_model_from_metadata = (
+                        await self._resolve_base_model_from_checkpoint(checkpoint_entry)
+                    )
                    if base_model_from_metadata:
                        metadata["base_model"] = base_model_from_metadata

            tags = self._parse_tags(params.get("tags"))
-            
+
            # 3. Download Image
-            image_bytes, extension, civitai_meta_from_download = await self._download_remote_media(image_url)
+            (
+                image_bytes,
+                extension,
+                civitai_meta_from_download,
+            ) = await self._download_remote_media(image_url)

            # 4. Extract Embedded Metadata
-            # Note: We still extract this here because Enricher currently expects 'gen_params' to already be populated 
-            # with embedded data if we want it to merge it. 
+            # Note: We still extract this here because Enricher currently expects 'gen_params' to already be populated
+            # with embedded data if we want it to merge it.
            # However, logic in Enricher merges: request > civitai > embedded.
-            # So we should gather embedded params and put them into the recipe's gen_params (as initial state) 
+            # So we should gather embedded params and put them into the recipe's gen_params (as initial state)
            # OR pass them to enricher to handle?
            # The interface of Enricher.enrich_recipe takes `recipe` (with gen_params) and `request_params`.
            # So let's extract embedded and put it into recipe['gen_params'] but careful not to overwrite request params.
            # Actually, `GenParamsMerger` which `Enricher` uses handles 3 layers.
            # But `Enricher` interface is: recipe['gen_params'] (as embedded) + request_params + civitai (fetched internally).
-            # Wait, `Enricher` fetches Civitai info internally based on URL. 
+            # Wait, `Enricher` fetches Civitai info internally based on URL.
            # `civitai_meta_from_download` is returned by `_download_remote_media` which might be useful if URL didn't have ID.
-            
+
            # Let's extract embedded metadata first
            embedded_gen_params = {}
            try:
-                with tempfile.NamedTemporaryFile(suffix=extension, delete=False) as temp_img:
+                with tempfile.NamedTemporaryFile(
+                    suffix=extension, delete=False
+                ) as temp_img:
                    temp_img.write(image_bytes)
                    temp_img_path = temp_img.name
-                
+
                try:
                    raw_embedded = ExifUtils.extract_image_metadata(temp_img_path)
                    if raw_embedded:
-                        parser = self._analysis_service._recipe_parser_factory.create_parser(raw_embedded)
+                        parser = (
+                            self._analysis_service._recipe_parser_factory.create_parser(
+                                raw_embedded
+                            )
+                        )
                        if parser:
-                            parsed_embedded = await parser.parse_metadata(raw_embedded, recipe_scanner=recipe_scanner)
+                            parsed_embedded = await parser.parse_metadata(
+                                raw_embedded, recipe_scanner=recipe_scanner
+                            )
                            if parsed_embedded and "gen_params" in parsed_embedded:
                                embedded_gen_params = parsed_embedded["gen_params"]
                        else:
@@ -724,7 +808,9 @@ class RecipeManagementHandler:
                    if os.path.exists(temp_img_path):
                        os.unlink(temp_img_path)
            except Exception as exc:
-                self._logger.warning("Failed to extract embedded metadata during import: %s", exc)
+                self._logger.warning(
+                    "Failed to extract embedded metadata during import: %s", exc
+                )

            # Pre-populate gen_params with embedded data so Enricher treats it as the "base" layer
            if embedded_gen_params:
@@ -732,18 +818,18 @@ class RecipeManagementHandler:
                # But wait, we want request params to override everything.
                # So we should set recipe['gen_params'] = embedded, and pass request params to enricher.
                metadata["gen_params"] = embedded_gen_params
-            
+
            # 5. Enrich with unified logic
            # This will fetch Civitai info (if URL matches) and merge: request > civitai > embedded
            civitai_client = self._civitai_client_getter()
            await RecipeEnricher.enrich_recipe(
-                recipe=metadata, 
+                recipe=metadata,
                civitai_client=civitai_client,
-                request_params=gen_params_request # Pass explicit request params here to override
+                request_params=gen_params_request,  # Pass explicit request params here to override
            )
-            
+
            # If we got civitai_meta from download but Enricher didn't fetch it (e.g. not a civitai URL or failed),
-            # we might want to manually merge it? 
+            # we might want to manually merge it?
            # But usually `import_remote_recipe` is used with Civitai URLs.
            # For now, relying on Enricher's internal fetch is consistent with repair.

@@ -762,7 +848,9 @@ class RecipeManagementHandler:
        except RecipeDownloadError as exc:
            return web.json_response({"error": str(exc)}, status=400)
        except Exception as exc:
-            self._logger.error("Error importing recipe from remote source: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error importing recipe from remote source: %s", exc, exc_info=True
+            )
            return web.json_response({"error": str(exc)}, status=500)

    async def delete_recipe(self, request: web.Request) -> web.Response:
@@ -816,7 +904,11 @@ class RecipeManagementHandler:
            target_path = data.get("target_path")
            if not recipe_id or not target_path:
                return web.json_response(
-                    {"success": False, "error": "recipe_id and target_path are required"}, status=400
+                    {
+                        "success": False,
+                        "error": "recipe_id and target_path are required",
+                    },
+                    status=400,
                )

            result = await self._persistence_service.move_recipe(
@@ -845,7 +937,11 @@ class RecipeManagementHandler:
            target_path = data.get("target_path")
            if not recipe_ids or not target_path:
                return web.json_response(
-                    {"success": False, "error": "recipe_ids and target_path are required"}, status=400
+                    {
+                        "success": False,
+                        "error": "recipe_ids and target_path are required",
+                    },
+                    status=400,
                )

            result = await self._persistence_service.move_recipes_bulk(
@@ -934,7 +1030,9 @@ class RecipeManagementHandler:
        except RecipeValidationError as exc:
            return web.json_response({"error": str(exc)}, status=400)
        except Exception as exc:
-            self._logger.error("Error saving recipe from widget: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error saving recipe from widget: %s", exc, exc_info=True
+            )
            return web.json_response({"error": str(exc)}, status=500)

    async def _parse_save_payload(self, reader) -> dict[str, Any]:
@@ -1006,7 +1104,9 @@ class RecipeManagementHandler:
            raise RecipeValidationError("gen_params payload must be an object")
        return parsed

-    def _parse_resources_payload(self, payload_raw: str) -> tuple[Optional[Dict[str, Any]], List[Dict[str, Any]]]:
+    def _parse_resources_payload(
+        self, payload_raw: str
+    ) -> tuple[Optional[Dict[str, Any]], List[Dict[str, Any]]]:
        try:
            payload = json.loads(payload_raw)
        except json.JSONDecodeError as exc:
@@ -1066,15 +1166,19 @@ class RecipeManagementHandler:
            civitai_match = re.match(r"https://civitai\.com/images/(\d+)", image_url)
            if civitai_match:
                if civitai_client is None:
-                    raise RecipeDownloadError("Civitai client unavailable for image download")
+                    raise RecipeDownloadError(
+                        "Civitai client unavailable for image download"
+                    )
                image_info = await civitai_client.get_image_info(civitai_match.group(1))
                if not image_info:
-                    raise RecipeDownloadError("Failed to fetch image information from Civitai")
-                
+                    raise RecipeDownloadError(
+                        "Failed to fetch image information from Civitai"
+                    )
+
                media_url = image_info.get("url")
                if not media_url:
                    raise RecipeDownloadError("No image URL found in Civitai response")
-                
+
                # Use optimized preview URLs if possible
                media_type = image_info.get("type")
                rewritten_url, _ = rewrite_preview_url(media_url, media_type=media_type)
@@ -1083,18 +1187,24 @@ class RecipeManagementHandler:
                else:
                    download_url = media_url

-            success, result = await downloader.download_file(download_url, temp_path, use_auth=False)
+            success, result = await downloader.download_file(
+                download_url, temp_path, use_auth=False
+            )
            if not success:
                raise RecipeDownloadError(f"Failed to download image: {result}")
-            
+
            # Extract extension from URL
-            url_path = download_url.split('?')[0].split('#')[0]
+            url_path = download_url.split("?")[0].split("#")[0]
            extension = os.path.splitext(url_path)[1].lower()
            if not extension:
-                extension = ".webp" # Default to webp if unknown
+                extension = ".webp"  # Default to webp if unknown

            with open(temp_path, "rb") as file_obj:
-                return file_obj.read(), extension, image_info.get("meta") if civitai_match and image_info else None
+                return (
+                    file_obj.read(),
+                    extension,
+                    image_info.get("meta") if civitai_match and image_info else None,
+                )
        except RecipeDownloadError:
            raise
        except RecipeValidationError:
@@ -1108,14 +1218,15 @@ class RecipeManagementHandler:
                except FileNotFoundError:
                    pass

-
    def _safe_int(self, value: Any) -> int:
        try:
            return int(value)
        except (TypeError, ValueError):
            return 0

-    async def _resolve_base_model_from_checkpoint(self, checkpoint_entry: Dict[str, Any]) -> str:
+    async def _resolve_base_model_from_checkpoint(
+        self, checkpoint_entry: Dict[str, Any]
+    ) -> str:
        version_id = self._safe_int(checkpoint_entry.get("modelVersionId"))

        if not version_id:
@@ -1134,7 +1245,9 @@ class RecipeManagementHandler:
                base_model = version_info.get("baseModel") or ""
                return str(base_model) if base_model is not None else ""
        except Exception as exc:  # pragma: no cover - defensive logging
-            self._logger.warning("Failed to resolve base model from checkpoint metadata: %s", exc)
+            self._logger.warning(
+                "Failed to resolve base model from checkpoint metadata: %s", exc
+            )

        return ""

@@ -1279,5 +1392,311 @@ class RecipeSharingHandler:
        except RecipeNotFoundError as exc:
            return web.json_response({"error": str(exc)}, status=404)
        except Exception as exc:
-            self._logger.error("Error downloading shared recipe: %s", exc, exc_info=True)
+            self._logger.error(
+                "Error downloading shared recipe: %s", exc, exc_info=True
+            )
            return web.json_response({"error": str(exc)}, status=500)
+
+
+class BatchImportHandler:
+    """Handle batch import operations for recipes."""
+
+    def __init__(
+        self,
+        *,
+        ensure_dependencies_ready: EnsureDependenciesCallable,
+        recipe_scanner_getter: RecipeScannerGetter,
+        civitai_client_getter: CivitaiClientGetter,
+        logger: Logger,
+        batch_import_service: BatchImportService,
+    ) -> None:
+        self._ensure_dependencies_ready = ensure_dependencies_ready
+        self._recipe_scanner_getter = recipe_scanner_getter
+        self._civitai_client_getter = civitai_client_getter
+        self._logger = logger
+        self._batch_import_service = batch_import_service
+
+    async def start_batch_import(self, request: web.Request) -> web.Response:
+        try:
+            await self._ensure_dependencies_ready()
+
+            if self._batch_import_service.is_import_running():
+                return web.json_response(
+                    {"success": False, "error": "Batch import already in progress"},
+                    status=409,
+                )
+
+            data = await request.json()
+            items = data.get("items", [])
+            tags = data.get("tags", [])
+            skip_no_metadata = data.get("skip_no_metadata", False)
+
+            if not items:
+                return web.json_response(
+                    {"success": False, "error": "No items provided"},
+                    status=400,
+                )
+
+            for item in items:
+                if not item.get("source"):
+                    return web.json_response(
+                        {
+                            "success": False,
+                            "error": "Each item must have a 'source' field",
+                        },
+                        status=400,
+                    )
+
+            operation_id = await self._batch_import_service.start_batch_import(
+                recipe_scanner_getter=self._recipe_scanner_getter,
+                civitai_client_getter=self._civitai_client_getter,
+                items=items,
+                tags=tags,
+                skip_no_metadata=skip_no_metadata,
+            )
+
+            return web.json_response(
+                {
+                    "success": True,
+                    "operation_id": operation_id,
+                }
+            )
+        except RecipeValidationError as exc:
+            return web.json_response({"success": False, "error": str(exc)}, status=400)
+        except Exception as exc:
+            self._logger.error("Error starting batch import: %s", exc, exc_info=True)
+            return web.json_response({"success": False, "error": str(exc)}, status=500)
+
+    async def start_directory_import(self, request: web.Request) -> web.Response:
+        try:
+            await self._ensure_dependencies_ready()
+
+            if self._batch_import_service.is_import_running():
+                return web.json_response(
+                    {"success": False, "error": "Batch import already in progress"},
+                    status=409,
+                )
+
+            data = await request.json()
+            directory = data.get("directory")
+            recursive = data.get("recursive", True)
+            tags = data.get("tags", [])
+            skip_no_metadata = data.get("skip_no_metadata", True)
+
+            if not directory:
+                return web.json_response(
+                    {"success": False, "error": "Directory path is required"},
+                    status=400,
+                )
+
+            operation_id = await self._batch_import_service.start_directory_import(
+                recipe_scanner_getter=self._recipe_scanner_getter,
+                civitai_client_getter=self._civitai_client_getter,
+                directory=directory,
+                recursive=recursive,
+                tags=tags,
+                skip_no_metadata=skip_no_metadata,
+            )
+
+            return web.json_response(
+                {
+                    "success": True,
+                    "operation_id": operation_id,
+                }
+            )
+        except RecipeValidationError as exc:
+            return web.json_response({"success": False, "error": str(exc)}, status=400)
+        except Exception as exc:
+            self._logger.error(
+                "Error starting directory import: %s", exc, exc_info=True
+            )
+            return web.json_response({"success": False, "error": str(exc)}, status=500)
+
+    async def get_batch_import_progress(self, request: web.Request) -> web.Response:
+        try:
+            operation_id = request.query.get("operation_id")
+            if not operation_id:
+                return web.json_response(
+                    {"success": False, "error": "operation_id is required"},
+                    status=400,
+                )
+
+            progress = self._batch_import_service.get_progress(operation_id)
+            if not progress:
+                return web.json_response(
+                    {"success": False, "error": "Operation not found"},
+                    status=404,
+                )
+
+            return web.json_response(
+                {
+                    "success": True,
+                    "progress": progress.to_dict(),
+                }
+            )
+        except Exception as exc:
+            self._logger.error(
+                "Error getting batch import progress: %s", exc, exc_info=True
+            )
+            return web.json_response({"success": False, "error": str(exc)}, status=500)
+
+    async def cancel_batch_import(self, request: web.Request) -> web.Response:
+        try:
+            data = await request.json()
+            operation_id = data.get("operation_id")
+
+            if not operation_id:
+                return web.json_response(
+                    {"success": False, "error": "operation_id is required"},
+                    status=400,
+                )
+
+            cancelled = self._batch_import_service.cancel_import(operation_id)
+            if not cancelled:
+                return web.json_response(
+                    {
+                        "success": False,
+                        "error": "Operation not found or already completed",
+                    },
+                    status=404,
+                )
+
+            return web.json_response(
+                {"success": True, "message": "Cancellation requested"}
+            )
+        except Exception as exc:
+            self._logger.error("Error cancelling batch import: %s", exc, exc_info=True)
+            return web.json_response({"success": False, "error": str(exc)}, status=500)
+
+    async def browse_directory(self, request: web.Request) -> web.Response:
+        """Browse a directory and return its contents (subdirectories and files)."""
+        try:
+            data = await request.json()
+            directory_path = data.get("path", "")
+
+            if not directory_path:
+                return web.json_response(
+                    {"success": False, "error": "Directory path is required"},
+                    status=400,
+                )
+
+            # Normalize the path
+            path = Path(directory_path).expanduser().resolve()
+
+            # Security check: ensure path is within allowed directories
+            # Allow common image/model directories
+            allowed_roots = [
+                Path.home(),
+                Path("/"),  # Allow browsing from root for flexibility
+            ]
+
+            # Check if path is within any allowed root
+            is_allowed = False
+            for root in allowed_roots:
+                try:
+                    path.relative_to(root)
+                    is_allowed = True
+                    break
+                except ValueError:
+                    continue
+
+            if not is_allowed:
+                return web.json_response(
+                    {"success": False, "error": "Access denied to this directory"},
+                    status=403,
+                )
+
+            if not path.exists():
+                return web.json_response(
+                    {"success": False, "error": "Directory does not exist"},
+                    status=404,
+                )
+
+            if not path.is_dir():
+                return web.json_response(
+                    {"success": False, "error": "Path is not a directory"},
+                    status=400,
+                )
+
+            # List directory contents
+            directories = []
+            image_files = []
+
+            image_extensions = {
+                ".jpg",
+                ".jpeg",
+                ".png",
+                ".gif",
+                ".webp",
+                ".bmp",
+                ".tiff",
+                ".tif",
+            }
+
+            try:
+                for item in path.iterdir():
+                    try:
+                        if item.is_dir():
+                            # Skip hidden directories and common system folders
+                            if not item.name.startswith(".") and item.name not in [
+                                "__pycache__",
+                                "node_modules",
+                            ]:
+                                directories.append(
+                                    {
+                                        "name": item.name,
+                                        "path": str(item),
+                                        "is_parent": False,
+                                    }
+                                )
+                        elif item.is_file() and item.suffix.lower() in image_extensions:
+                            image_files.append(
+                                {
+                                    "name": item.name,
+                                    "path": str(item),
+                                    "size": item.stat().st_size,
+                                }
+                            )
+                    except (PermissionError, OSError):
+                        # Skip files/directories we can't access
+                        continue
+
+                # Sort directories and files alphabetically
+                directories.sort(key=lambda x: x["name"].lower())
+                image_files.sort(key=lambda x: x["name"].lower())
+
+                # Add parent directory if not at root
+                parent_path = path.parent
+                show_parent = str(path) != str(path.root)
+
+                return web.json_response(
+                    {
+                        "success": True,
+                        "current_path": str(path),
+                        "parent_path": str(parent_path) if show_parent else None,
+                        "directories": directories,
+                        "image_files": image_files,
+                        "image_count": len(image_files),
+                        "directory_count": len(directories),
+                    }
+                )
+
+            except PermissionError:
+                return web.json_response(
+                    {"success": False, "error": "Permission denied"},
+                    status=403,
+                )
+            except OSError as exc:
+                return web.json_response(
+                    {"success": False, "error": f"Error reading directory: {str(exc)}"},
+                    status=500,
+                )
+
+        except json.JSONDecodeError:
+            return web.json_response(
+                {"success": False, "error": "Invalid JSON"},
+                status=400,
+            )
+        except Exception as exc:
+            self._logger.error("Error browsing directory: %s", exc, exc_info=True)
+            return web.json_response({"success": False, "error": str(exc)}, status=500)
--- a/py/routes/misc_route_registrar.py
+++ b/py/routes/misc_route_registrar.py
@@ -38,12 +38,24 @@ MISC_ROUTE_DEFINITIONS: tuple[RouteDefinition, ...] = (
    RouteDefinition("GET", "/api/lm/get-registry", "get_registry"),
    RouteDefinition("GET", "/api/lm/check-model-exists", "check_model_exists"),
    RouteDefinition("GET", "/api/lm/civitai/user-models", "get_civitai_user_models"),
-    RouteDefinition("POST", "/api/lm/download-metadata-archive", "download_metadata_archive"),
-    RouteDefinition("POST", "/api/lm/remove-metadata-archive", "remove_metadata_archive"),
-    RouteDefinition("GET", "/api/lm/metadata-archive-status", "get_metadata_archive_status"),
-    RouteDefinition("GET", "/api/lm/model-versions-status", "get_model_versions_status"),
+    RouteDefinition(
+        "POST", "/api/lm/download-metadata-archive", "download_metadata_archive"
+    ),
+    RouteDefinition(
+        "POST", "/api/lm/remove-metadata-archive", "remove_metadata_archive"
+    ),
+    RouteDefinition(
+        "GET", "/api/lm/metadata-archive-status", "get_metadata_archive_status"
+    ),
+    RouteDefinition(
+        "GET", "/api/lm/model-versions-status", "get_model_versions_status"
+    ),
    RouteDefinition("POST", "/api/lm/settings/open-location", "open_settings_location"),
    RouteDefinition("GET", "/api/lm/custom-words/search", "search_custom_words"),
+    RouteDefinition("GET", "/api/lm/example-workflows", "get_example_workflows"),
+    RouteDefinition(
+        "GET", "/api/lm/example-workflows/{filename}", "get_example_workflow"
+    ),
 )


@@ -67,7 +79,11 @@ class MiscRouteRegistrar:
        definitions: Iterable[RouteDefinition] = MISC_ROUTE_DEFINITIONS,
    ) -> None:
        for definition in definitions:
-            self._bind(definition.method, definition.path, handler_lookup[definition.handler_name])
+            self._bind(
+                definition.method,
+                definition.path,
+                handler_lookup[definition.handler_name],
+            )

    def _bind(self, method: str, path: str, handler: Callable) -> None:
        add_method_name = self._METHOD_MAP[method.upper()]
--- a/py/routes/misc_routes.py
+++ b/py/routes/misc_routes.py
@@ -19,6 +19,7 @@ from ..services.downloader import get_downloader
 from ..utils.usage_stats import UsageStats
 from .handlers.misc_handlers import (
    CustomWordsHandler,
+    ExampleWorkflowsHandler,
    FileSystemHandler,
    HealthCheckHandler,
    LoraCodeHandler,
@@ -38,9 +39,10 @@ from .misc_route_registrar import MiscRouteRegistrar

 logger = logging.getLogger(__name__)

-standalone_mode = os.environ.get("LORA_MANAGER_STANDALONE", "0") == "1" or os.environ.get(
-    "HF_HUB_DISABLE_TELEMETRY", "0"
-) == "0"
+standalone_mode = (
+    os.environ.get("LORA_MANAGER_STANDALONE", "0") == "1"
+    or os.environ.get("HF_HUB_DISABLE_TELEMETRY", "0") == "0"
+)


 class MiscRoutes:
@@ -75,7 +77,9 @@ class MiscRoutes:
        self._node_registry = node_registry or NodeRegistry()
        self._standalone_mode = standalone_mode_flag

-        self._handler_mapping: Mapping[str, Callable[[web.Request], Awaitable[web.StreamResponse]]] | None = None
+        self._handler_mapping: (
+            Mapping[str, Callable[[web.Request], Awaitable[web.StreamResponse]]] | None
+        ) = None

    @staticmethod
    def setup_routes(app: web.Application) -> None:
@@ -87,7 +91,9 @@ class MiscRoutes:
        registrar = self._registrar_factory(app)
        registrar.register_routes(self._ensure_handler_mapping())

-    def _ensure_handler_mapping(self) -> Mapping[str, Callable[[web.Request], Awaitable[web.StreamResponse]]]:
+    def _ensure_handler_mapping(
+        self,
+    ) -> Mapping[str, Callable[[web.Request], Awaitable[web.StreamResponse]]]:
        if self._handler_mapping is None:
            handler_set = self._create_handler_set()
            self._handler_mapping = handler_set.to_route_mapping()
@@ -121,6 +127,7 @@ class MiscRoutes:
        )
        custom_words = CustomWordsHandler()
        supporters = SupportersHandler()
+        example_workflows = ExampleWorkflowsHandler()

        return self._handler_set_factory(
            health=health,
@@ -135,6 +142,7 @@ class MiscRoutes:
            filesystem=filesystem,
            custom_words=custom_words,
            supporters=supporters,
+            example_workflows=example_workflows,
        )


--- a/py/routes/recipe_route_registrar.py
+++ b/py/routes/recipe_route_registrar.py
@@ -1,4 +1,5 @@
 """Route registrar for recipe endpoints."""
+
 from __future__ import annotations

 from dataclasses import dataclass
@@ -22,7 +23,9 @@ ROUTE_DEFINITIONS: tuple[RouteDefinition, ...] = (
    RouteDefinition("GET", "/api/lm/recipe/{recipe_id}", "get_recipe"),
    RouteDefinition("GET", "/api/lm/recipes/import-remote", "import_remote_recipe"),
    RouteDefinition("POST", "/api/lm/recipes/analyze-image", "analyze_uploaded_image"),
-    RouteDefinition("POST", "/api/lm/recipes/analyze-local-image", "analyze_local_image"),
+    RouteDefinition(
+        "POST", "/api/lm/recipes/analyze-local-image", "analyze_local_image"
+    ),
    RouteDefinition("POST", "/api/lm/recipes/save", "save_recipe"),
    RouteDefinition("DELETE", "/api/lm/recipe/{recipe_id}", "delete_recipe"),
    RouteDefinition("GET", "/api/lm/recipes/top-tags", "get_top_tags"),
@@ -30,9 +33,13 @@ ROUTE_DEFINITIONS: tuple[RouteDefinition, ...] = (
    RouteDefinition("GET", "/api/lm/recipes/roots", "get_roots"),
    RouteDefinition("GET", "/api/lm/recipes/folders", "get_folders"),
    RouteDefinition("GET", "/api/lm/recipes/folder-tree", "get_folder_tree"),
-    RouteDefinition("GET", "/api/lm/recipes/unified-folder-tree", "get_unified_folder_tree"),
+    RouteDefinition(
+        "GET", "/api/lm/recipes/unified-folder-tree", "get_unified_folder_tree"
+    ),
    RouteDefinition("GET", "/api/lm/recipe/{recipe_id}/share", "share_recipe"),
-    RouteDefinition("GET", "/api/lm/recipe/{recipe_id}/share/download", "download_shared_recipe"),
+    RouteDefinition(
+        "GET", "/api/lm/recipe/{recipe_id}/share/download", "download_shared_recipe"
+    ),
    RouteDefinition("GET", "/api/lm/recipe/{recipe_id}/syntax", "get_recipe_syntax"),
    RouteDefinition("PUT", "/api/lm/recipe/{recipe_id}/update", "update_recipe"),
    RouteDefinition("POST", "/api/lm/recipe/move", "move_recipe"),
@@ -40,13 +47,26 @@ ROUTE_DEFINITIONS: tuple[RouteDefinition, ...] = (
    RouteDefinition("POST", "/api/lm/recipe/lora/reconnect", "reconnect_lora"),
    RouteDefinition("GET", "/api/lm/recipes/find-duplicates", "find_duplicates"),
    RouteDefinition("POST", "/api/lm/recipes/bulk-delete", "bulk_delete"),
-    RouteDefinition("POST", "/api/lm/recipes/save-from-widget", "save_recipe_from_widget"),
+    RouteDefinition(
+        "POST", "/api/lm/recipes/save-from-widget", "save_recipe_from_widget"
+    ),
    RouteDefinition("GET", "/api/lm/recipes/for-lora", "get_recipes_for_lora"),
    RouteDefinition("GET", "/api/lm/recipes/scan", "scan_recipes"),
    RouteDefinition("POST", "/api/lm/recipes/repair", "repair_recipes"),
    RouteDefinition("POST", "/api/lm/recipes/cancel-repair", "cancel_repair"),
    RouteDefinition("POST", "/api/lm/recipe/{recipe_id}/repair", "repair_recipe"),
    RouteDefinition("GET", "/api/lm/recipes/repair-progress", "get_repair_progress"),
+    RouteDefinition("POST", "/api/lm/recipes/batch-import/start", "start_batch_import"),
+    RouteDefinition(
+        "GET", "/api/lm/recipes/batch-import/progress", "get_batch_import_progress"
+    ),
+    RouteDefinition(
+        "POST", "/api/lm/recipes/batch-import/cancel", "cancel_batch_import"
+    ),
+    RouteDefinition(
+        "POST", "/api/lm/recipes/batch-import/directory", "start_directory_import"
+    ),
+    RouteDefinition("POST", "/api/lm/recipes/browse-directory", "browse_directory"),
 )


@@ -63,7 +83,9 @@ class RecipeRouteRegistrar:
    def __init__(self, app: web.Application) -> None:
        self._app = app

-    def register_routes(self, handler_lookup: Mapping[str, Callable[[web.Request], object]]) -> None:
+    def register_routes(
+        self, handler_lookup: Mapping[str, Callable[[web.Request], object]]
+    ) -> None:
        for definition in ROUTE_DEFINITIONS:
            handler = handler_lookup[definition.handler_name]
            self._bind_route(definition.method, definition.path, handler)
--- a/py/services/base_model_service.py
+++ b/py/services/base_model_service.py
@@ -1,5 +1,6 @@
 from abc import ABC, abstractmethod
 import asyncio
+import re
 from typing import Any, Dict, List, Optional, Type, TYPE_CHECKING
 import logging
 import os
@@ -207,7 +208,11 @@ class BaseModelService(ABC):

        reverse = sort_params.order == "desc"
        annotated.sort(
-            key=lambda x: (x.get("usage_count", 0), x.get("model_name", "").lower()),
+            key=lambda x: (
+                x.get("usage_count", 0),
+                x.get("model_name", "").lower(),
+                x.get("file_path", "").lower()
+            ),
            reverse=reverse,
        )
        return annotated
@@ -383,7 +388,9 @@ class BaseModelService(ABC):
        # Check user setting for hiding early access updates
        hide_early_access = False
        try:
-            hide_early_access = bool(self.settings.get("hide_early_access_updates", False))
+            hide_early_access = bool(
+                self.settings.get("hide_early_access_updates", False)
+            )
        except Exception:
            hide_early_access = False

@@ -413,7 +420,11 @@ class BaseModelService(ABC):
            bulk_method = getattr(self.update_service, "has_updates_bulk", None)
            if callable(bulk_method):
                try:
-                    resolved = await bulk_method(self.model_type, ordered_ids, hide_early_access=hide_early_access)
+                    resolved = await bulk_method(
+                        self.model_type,
+                        ordered_ids,
+                        hide_early_access=hide_early_access,
+                    )
                except Exception as exc:
                    logger.error(
                        "Failed to resolve update status in bulk for %s models (%s): %s",
@@ -426,7 +437,9 @@ class BaseModelService(ABC):

        if resolved is None:
            tasks = [
-                self.update_service.has_update(self.model_type, model_id, hide_early_access=hide_early_access)
+                self.update_service.has_update(
+                    self.model_type, model_id, hide_early_access=hide_early_access
+                )
                for model_id in ordered_ids
            ]
            results = await asyncio.gather(*tasks, return_exceptions=True)
@@ -588,13 +601,19 @@ class BaseModelService(ABC):
            normalized_type = normalize_sub_type(resolve_sub_type(entry))
            if not normalized_type:
                continue
-            
+
            # Filter by valid sub-types based on scanner type
-            if self.model_type == "lora" and normalized_type not in VALID_LORA_SUB_TYPES:
+            if (
+                self.model_type == "lora"
+                and normalized_type not in VALID_LORA_SUB_TYPES
+            ):
                continue
-            if self.model_type == "checkpoint" and normalized_type not in VALID_CHECKPOINT_SUB_TYPES:
+            if (
+                self.model_type == "checkpoint"
+                and normalized_type not in VALID_CHECKPOINT_SUB_TYPES
+            ):
                continue
-            
+
            type_counts[normalized_type] = type_counts.get(normalized_type, 0) + 1

        sorted_types = sorted(
@@ -807,38 +826,61 @@ class BaseModelService(ABC):

        return include_terms, exclude_terms

+    @staticmethod
+    def _remove_model_extension(path: str) -> str:
+        """Remove model file extension (.safetensors, .ckpt, .pt, .bin) for cleaner matching."""
+        return re.sub(r"\.(safetensors|ckpt|pt|bin)$", "", path, flags=re.IGNORECASE)
+
    @staticmethod
    def _relative_path_matches_tokens(
        path_lower: str, include_terms: List[str], exclude_terms: List[str]
    ) -> bool:
-        """Determine whether a relative path string satisfies include/exclude tokens."""
-        if any(term and term in path_lower for term in exclude_terms):
+        """Determine whether a relative path string satisfies include/exclude tokens.
+
+        Matches against the path without extension to avoid matching .safetensors
+        when searching for 's'.
+        """
+        # Use path without extension for matching
+        path_for_matching = BaseModelService._remove_model_extension(path_lower)
+
+        if any(term and term in path_for_matching for term in exclude_terms):
            return False

        for term in include_terms:
-            if term and term not in path_lower:
+            if term and term not in path_for_matching:
                return False

        return True

    @staticmethod
    def _relative_path_sort_key(relative_path: str, include_terms: List[str]) -> tuple:
-        """Sort paths by how well they satisfy the include tokens."""
-        path_lower = relative_path.lower()
+        """Sort paths by how well they satisfy the include tokens.
+
+        Sorts based on path without extension for consistent ordering.
+        """
+        # Use path without extension for sorting
+        path_for_sorting = BaseModelService._remove_model_extension(
+            relative_path.lower()
+        )
        prefix_hits = sum(
-            1 for term in include_terms if term and path_lower.startswith(term)
+            1 for term in include_terms if term and path_for_sorting.startswith(term)
        )
        match_positions = [
-            path_lower.find(term)
+            path_for_sorting.find(term)
            for term in include_terms
-            if term and term in path_lower
+            if term and term in path_for_sorting
        ]
        first_match_index = min(match_positions) if match_positions else 0

-        return (-prefix_hits, first_match_index, len(relative_path), path_lower)
+        return (
+            -prefix_hits,
+            first_match_index,
+            len(path_for_sorting),
+            path_for_sorting,
+        )

    async def search_relative_paths(
-        self, search_term: str, limit: int = 15
+        self, search_term: str, limit: int = 15, offset: int = 0
    ) -> List[str]:
        """Search model relative file paths for autocomplete functionality"""
        cache = await self.scanner.get_cached_data()
@@ -849,6 +891,7 @@ class BaseModelService(ABC):
        # Get model roots for path calculation
        model_roots = self.scanner.get_model_roots()

+        # Collect all matching paths first (needed for proper sorting and offset)
        for model in cache.raw_data:
            file_path = model.get("file_path", "")
            if not file_path:
@@ -877,12 +920,12 @@ class BaseModelService(ABC):
            ):
                matching_paths.append(relative_path)

-                if len(matching_paths) >= limit * 2:  # Get more for better sorting
-                    break
-
        # Sort by relevance (prefix and earliest hits first, then by length and alphabetically)
        matching_paths.sort(
            key=lambda relative: self._relative_path_sort_key(relative, include_terms)
        )

-        return matching_paths[:limit]
+        # Apply offset and limit
+        start = min(offset, len(matching_paths))
+        end = min(start + limit, len(matching_paths))
+        return matching_paths[start:end]
--- a/py/services/batch_import_service.py
+++ b/py/services/batch_import_service.py
@@ -0,0 +1,593 @@
+"""Batch import service for importing multiple images as recipes."""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+import os
+import time
+import uuid
+from dataclasses import dataclass, field
+from enum import Enum
+from typing import Any, Callable, Dict, List, Optional, Set
+
+from aiohttp import web
+
+from .recipes import (
+    RecipeAnalysisService,
+    RecipePersistenceService,
+    RecipeValidationError,
+    RecipeDownloadError,
+    RecipeNotFoundError,
+)
+
+
+class ImportItemType(Enum):
+    """Type of import item."""
+
+    URL = "url"
+    LOCAL_PATH = "local_path"
+
+
+class ImportStatus(Enum):
+    """Status of an individual import item."""
+
+    PENDING = "pending"
+    PROCESSING = "processing"
+    SUCCESS = "success"
+    FAILED = "failed"
+    SKIPPED = "skipped"
+
+
+@dataclass
+class BatchImportItem:
+    """Represents a single item to import."""
+
+    id: str
+    source: str
+    item_type: ImportItemType
+    status: ImportStatus = ImportStatus.PENDING
+    error_message: Optional[str] = None
+    recipe_name: Optional[str] = None
+    recipe_id: Optional[str] = None
+    duration: float = 0.0
+
+
+@dataclass
+class BatchImportProgress:
+    """Tracks progress of a batch import operation."""
+
+    operation_id: str
+    total: int
+    completed: int = 0
+    success: int = 0
+    failed: int = 0
+    skipped: int = 0
+    current_item: str = ""
+    status: str = "pending"
+    started_at: float = field(default_factory=time.time)
+    finished_at: Optional[float] = None
+    items: List[BatchImportItem] = field(default_factory=list)
+    tags: List[str] = field(default_factory=list)
+    skip_no_metadata: bool = False
+    skip_duplicates: bool = False
+
+    def to_dict(self) -> Dict[str, Any]:
+        return {
+            "operation_id": self.operation_id,
+            "total": self.total,
+            "completed": self.completed,
+            "success": self.success,
+            "failed": self.failed,
+            "skipped": self.skipped,
+            "current_item": self.current_item,
+            "status": self.status,
+            "started_at": self.started_at,
+            "finished_at": self.finished_at,
+            "progress_percent": round((self.completed / self.total) * 100, 1)
+            if self.total > 0
+            else 0,
+            "items": [
+                {
+                    "id": item.id,
+                    "source": item.source,
+                    "item_type": item.item_type.value,
+                    "status": item.status.value,
+                    "error_message": item.error_message,
+                    "recipe_name": item.recipe_name,
+                    "recipe_id": item.recipe_id,
+                    "duration": item.duration,
+                }
+                for item in self.items
+            ],
+        }
+
+
+class AdaptiveConcurrencyController:
+    """Adjusts concurrency based on task performance."""
+
+    def __init__(
+        self,
+        min_concurrency: int = 1,
+        max_concurrency: int = 5,
+        initial_concurrency: int = 3,
+    ) -> None:
+        self.min_concurrency = min_concurrency
+        self.max_concurrency = max_concurrency
+        self.current_concurrency = initial_concurrency
+        self._task_durations: List[float] = []
+        self._recent_errors = 0
+        self._recent_successes = 0
+
+    def record_result(self, duration: float, success: bool) -> None:
+        self._task_durations.append(duration)
+        if len(self._task_durations) > 10:
+            self._task_durations.pop(0)
+
+        if success:
+            self._recent_successes += 1
+            if duration < 1.0 and self.current_concurrency < self.max_concurrency:
+                self.current_concurrency = min(
+                    self.current_concurrency + 1, self.max_concurrency
+                )
+            elif duration > 10.0 and self.current_concurrency > self.min_concurrency:
+                self.current_concurrency = max(
+                    self.current_concurrency - 1, self.min_concurrency
+                )
+        else:
+            self._recent_errors += 1
+            if self.current_concurrency > self.min_concurrency:
+                self.current_concurrency = max(
+                    self.current_concurrency - 1, self.min_concurrency
+                )
+
+    def reset_counters(self) -> None:
+        self._recent_errors = 0
+        self._recent_successes = 0
+
+    def get_semaphore(self) -> asyncio.Semaphore:
+        return asyncio.Semaphore(self.current_concurrency)
+
+
+class BatchImportService:
+    """Service for batch importing images as recipes."""
+
+    SUPPORTED_EXTENSIONS: Set[str] = {".jpg", ".jpeg", ".png", ".webp", ".gif", ".bmp"}
+
+    def __init__(
+        self,
+        *,
+        analysis_service: RecipeAnalysisService,
+        persistence_service: RecipePersistenceService,
+        ws_manager: Any,
+        logger: logging.Logger,
+    ) -> None:
+        self._analysis_service = analysis_service
+        self._persistence_service = persistence_service
+        self._ws_manager = ws_manager
+        self._logger = logger
+        self._active_operations: Dict[str, BatchImportProgress] = {}
+        self._cancellation_flags: Dict[str, bool] = {}
+        self._concurrency_controller = AdaptiveConcurrencyController()
+
+    def is_import_running(self, operation_id: Optional[str] = None) -> bool:
+        if operation_id:
+            progress = self._active_operations.get(operation_id)
+            return progress is not None and progress.status in ("pending", "running")
+        return any(
+            p.status in ("pending", "running") for p in self._active_operations.values()
+        )
+
+    def get_progress(self, operation_id: str) -> Optional[BatchImportProgress]:
+        return self._active_operations.get(operation_id)
+
+    def cancel_import(self, operation_id: str) -> bool:
+        if operation_id in self._active_operations:
+            self._cancellation_flags[operation_id] = True
+            return True
+        return False
+
+    def _validate_url(self, url: str) -> bool:
+        import re
+
+        url_pattern = re.compile(
+            r"^https?://"
+            r"(?:(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+[A-Z]{2,6}\.?|"
+            r"localhost|"
+            r"\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})"
+            r"(?::\d+)?"
+            r"(?:/?|[/?]\S+)$",
+            re.IGNORECASE,
+        )
+        return url_pattern.match(url) is not None
+
+    def _validate_local_path(self, path: str) -> bool:
+        try:
+            normalized = os.path.normpath(path)
+            if not os.path.isabs(normalized):
+                return False
+            if ".." in normalized:
+                return False
+            return True
+        except Exception:
+            return False
+
+    def _is_duplicate_source(
+        self,
+        source: str,
+        item_type: ImportItemType,
+        recipe_scanner: Any,
+    ) -> bool:
+        try:
+            cache = recipe_scanner.get_cached_data_sync()
+            if not cache:
+                return False
+
+            for recipe in getattr(cache, "raw_data", []):
+                source_path = recipe.get("source_path") or recipe.get("source_url")
+                if source_path and source_path == source:
+                    return True
+            return False
+        except Exception:
+            self._logger.warning("Failed to check for duplicates", exc_info=True)
+            return False
+
+    async def start_batch_import(
+        self,
+        *,
+        recipe_scanner_getter: Callable[[], Any],
+        civitai_client_getter: Callable[[], Any],
+        items: List[Dict[str, str]],
+        tags: Optional[List[str]] = None,
+        skip_no_metadata: bool = False,
+        skip_duplicates: bool = False,
+    ) -> str:
+        operation_id = str(uuid.uuid4())
+
+        import_items = []
+        for idx, item in enumerate(items):
+            source = item.get("source", "")
+            item_type_str = item.get("type", "url")
+
+            if item_type_str == "url" or source.startswith(("http://", "https://")):
+                item_type = ImportItemType.URL
+            else:
+                item_type = ImportItemType.LOCAL_PATH
+
+            batch_import_item = BatchImportItem(
+                id=f"{operation_id}_{idx}",
+                source=source,
+                item_type=item_type,
+            )
+            import_items.append(batch_import_item)
+
+        progress = BatchImportProgress(
+            operation_id=operation_id,
+            total=len(import_items),
+            items=import_items,
+            tags=tags or [],
+            skip_no_metadata=skip_no_metadata,
+            skip_duplicates=skip_duplicates,
+        )
+
+        self._active_operations[operation_id] = progress
+        self._cancellation_flags[operation_id] = False
+
+        asyncio.create_task(
+            self._run_batch_import(
+                operation_id=operation_id,
+                recipe_scanner_getter=recipe_scanner_getter,
+                civitai_client_getter=civitai_client_getter,
+            )
+        )
+
+        return operation_id
+
+    async def start_directory_import(
+        self,
+        *,
+        recipe_scanner_getter: Callable[[], Any],
+        civitai_client_getter: Callable[[], Any],
+        directory: str,
+        recursive: bool = True,
+        tags: Optional[List[str]] = None,
+        skip_no_metadata: bool = False,
+        skip_duplicates: bool = False,
+    ) -> str:
+        image_paths = await self._discover_images(directory, recursive)
+
+        items = [{"source": path, "type": "local_path"} for path in image_paths]
+
+        return await self.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=items,
+            tags=tags,
+            skip_no_metadata=skip_no_metadata,
+            skip_duplicates=skip_duplicates,
+        )
+
+    async def _discover_images(
+        self,
+        directory: str,
+        recursive: bool = True,
+    ) -> List[str]:
+        if not os.path.isdir(directory):
+            raise RecipeValidationError(f"Directory not found: {directory}")
+
+        image_paths: List[str] = []
+
+        if recursive:
+            for root, _, files in os.walk(directory):
+                for filename in files:
+                    if self._is_supported_image(filename):
+                        image_paths.append(os.path.join(root, filename))
+        else:
+            for filename in os.listdir(directory):
+                filepath = os.path.join(directory, filename)
+                if os.path.isfile(filepath) and self._is_supported_image(filename):
+                    image_paths.append(filepath)
+
+        return sorted(image_paths)
+
+    def _is_supported_image(self, filename: str) -> bool:
+        ext = os.path.splitext(filename)[1].lower()
+        return ext in self.SUPPORTED_EXTENSIONS
+
+    async def _run_batch_import(
+        self,
+        *,
+        operation_id: str,
+        recipe_scanner_getter: Callable[[], Any],
+        civitai_client_getter: Callable[[], Any],
+    ) -> None:
+        progress = self._active_operations.get(operation_id)
+        if not progress:
+            return
+
+        progress.status = "running"
+        await self._broadcast_progress(progress)
+
+        self._concurrency_controller = AdaptiveConcurrencyController()
+
+        async def process_item(item: BatchImportItem) -> None:
+            if self._cancellation_flags.get(operation_id, False):
+                return
+
+            progress.current_item = (
+                os.path.basename(item.source)
+                if item.item_type == ImportItemType.LOCAL_PATH
+                else item.source[:50]
+            )
+            item.status = ImportStatus.PROCESSING
+            await self._broadcast_progress(progress)
+
+            start_time = time.time()
+            try:
+                result = await self._import_single_item(
+                    item=item,
+                    recipe_scanner_getter=recipe_scanner_getter,
+                    civitai_client_getter=civitai_client_getter,
+                    tags=progress.tags,
+                    skip_no_metadata=progress.skip_no_metadata,
+                    skip_duplicates=progress.skip_duplicates,
+                    semaphore=self._concurrency_controller.get_semaphore(),
+                )
+
+                duration = time.time() - start_time
+                item.duration = duration
+                self._concurrency_controller.record_result(
+                    duration, result.get("success", False)
+                )
+
+                if result.get("success"):
+                    item.status = ImportStatus.SUCCESS
+                    item.recipe_name = result.get("recipe_name")
+                    item.recipe_id = result.get("recipe_id")
+                    progress.success += 1
+                elif result.get("skipped"):
+                    item.status = ImportStatus.SKIPPED
+                    item.error_message = result.get("error")
+                    progress.skipped += 1
+                else:
+                    item.status = ImportStatus.FAILED
+                    item.error_message = result.get("error")
+                    progress.failed += 1
+
+            except Exception as e:
+                self._logger.error(f"Error importing {item.source}: {e}")
+                item.status = ImportStatus.FAILED
+                item.error_message = str(e)
+                item.duration = time.time() - start_time
+                progress.failed += 1
+                self._concurrency_controller.record_result(item.duration, False)
+
+            progress.completed += 1
+            await self._broadcast_progress(progress)
+
+        tasks = [process_item(item) for item in progress.items]
+        await asyncio.gather(*tasks, return_exceptions=True)
+
+        if self._cancellation_flags.get(operation_id, False):
+            progress.status = "cancelled"
+        else:
+            progress.status = "completed"
+
+        progress.finished_at = time.time()
+        progress.current_item = ""
+        await self._broadcast_progress(progress)
+
+        await asyncio.sleep(5)
+        self._cleanup_operation(operation_id)
+
+    async def _import_single_item(
+        self,
+        *,
+        item: BatchImportItem,
+        recipe_scanner_getter: Callable[[], Any],
+        civitai_client_getter: Callable[[], Any],
+        tags: List[str],
+        skip_no_metadata: bool,
+        skip_duplicates: bool,
+        semaphore: asyncio.Semaphore,
+    ) -> Dict[str, Any]:
+        async with semaphore:
+            recipe_scanner = recipe_scanner_getter()
+            if recipe_scanner is None:
+                return {"success": False, "error": "Recipe scanner unavailable"}
+
+            try:
+                if item.item_type == ImportItemType.URL:
+                    if not self._validate_url(item.source):
+                        return {
+                            "success": False,
+                            "error": f"Invalid URL format: {item.source}",
+                        }
+
+                    if skip_duplicates:
+                        if self._is_duplicate_source(
+                            item.source, item.item_type, recipe_scanner
+                        ):
+                            return {
+                                "success": False,
+                                "skipped": True,
+                                "error": "Duplicate source URL",
+                            }
+
+                    civitai_client = civitai_client_getter()
+                    analysis_result = await self._analysis_service.analyze_remote_image(
+                        url=item.source,
+                        recipe_scanner=recipe_scanner,
+                        civitai_client=civitai_client,
+                    )
+                else:
+                    if not self._validate_local_path(item.source):
+                        return {
+                            "success": False,
+                            "error": f"Invalid or unsafe path: {item.source}",
+                        }
+
+                    if not os.path.exists(item.source):
+                        return {
+                            "success": False,
+                            "error": f"File not found: {item.source}",
+                        }
+
+                    if skip_duplicates:
+                        if self._is_duplicate_source(
+                            item.source, item.item_type, recipe_scanner
+                        ):
+                            return {
+                                "success": False,
+                                "skipped": True,
+                                "error": "Duplicate source path",
+                            }
+
+                    analysis_result = await self._analysis_service.analyze_local_image(
+                        file_path=item.source,
+                        recipe_scanner=recipe_scanner,
+                    )
+
+                payload = analysis_result.payload
+
+                if payload.get("error"):
+                    if skip_no_metadata and "No metadata" in payload.get("error", ""):
+                        return {
+                            "success": False,
+                            "skipped": True,
+                            "error": payload["error"],
+                        }
+                    return {"success": False, "error": payload["error"]}
+
+                loras = payload.get("loras", [])
+                if not loras:
+                    if skip_no_metadata:
+                        return {
+                            "success": False,
+                            "skipped": True,
+                            "error": "No LoRAs found in image",
+                        }
+                    # When skip_no_metadata is False, allow importing images without LoRAs
+                    # Continue with empty loras list
+
+                recipe_name = self._generate_recipe_name(item, payload)
+                all_tags = list(set(tags + (payload.get("tags", []) or [])))
+
+                metadata = {
+                    "base_model": payload.get("base_model", ""),
+                    "loras": loras,
+                    "gen_params": payload.get("gen_params", {}),
+                    "source_path": item.source,
+                }
+
+                if payload.get("checkpoint"):
+                    metadata["checkpoint"] = payload["checkpoint"]
+
+                image_bytes = None
+                image_base64 = payload.get("image_base64")
+
+                if item.item_type == ImportItemType.LOCAL_PATH:
+                    with open(item.source, "rb") as f:
+                        image_bytes = f.read()
+                    image_base64 = None
+
+                save_result = await self._persistence_service.save_recipe(
+                    recipe_scanner=recipe_scanner,
+                    image_bytes=image_bytes,
+                    image_base64=image_base64,
+                    name=recipe_name,
+                    tags=all_tags,
+                    metadata=metadata,
+                    extension=payload.get("extension"),
+                )
+
+                if save_result.status == 200:
+                    return {
+                        "success": True,
+                        "recipe_name": recipe_name,
+                        "recipe_id": save_result.payload.get("id"),
+                    }
+                else:
+                    return {
+                        "success": False,
+                        "error": save_result.payload.get(
+                            "error", "Failed to save recipe"
+                        ),
+                    }
+
+            except RecipeValidationError as e:
+                return {"success": False, "error": str(e)}
+            except RecipeDownloadError as e:
+                return {"success": False, "error": str(e)}
+            except RecipeNotFoundError as e:
+                return {"success": False, "skipped": True, "error": str(e)}
+            except Exception as e:
+                self._logger.error(
+                    f"Unexpected error importing {item.source}: {e}", exc_info=True
+                )
+                return {"success": False, "error": str(e)}
+
+    def _generate_recipe_name(
+        self, item: BatchImportItem, payload: Dict[str, Any]
+    ) -> str:
+        if item.item_type == ImportItemType.LOCAL_PATH:
+            base_name = os.path.splitext(os.path.basename(item.source))[0]
+            return base_name[:100]
+        else:
+            loras = payload.get("loras", [])
+            if loras:
+                first_lora = loras[0].get("name", "Recipe")
+                return f"Import - {first_lora}"[:100]
+            return f"Imported Recipe {item.id[:8]}"
+
+    async def _broadcast_progress(self, progress: BatchImportProgress) -> None:
+        await self._ws_manager.broadcast(
+            {
+                "type": "batch_import_progress",
+                **progress.to_dict(),
+            }
+        )
+
+    def _cleanup_operation(self, operation_id: str) -> None:
+        if operation_id in self._cancellation_flags:
+            del self._cancellation_flags[operation_id]
--- a/py/services/checkpoint_scanner.py
+++ b/py/services/checkpoint_scanner.py
@@ -13,22 +13,35 @@ from .model_hash_index import ModelHashIndex

 logger = logging.getLogger(__name__)

+
 class CheckpointScanner(ModelScanner):
    """Service for scanning and managing checkpoint files"""
-    
+
    def __init__(self):
        # Define supported file extensions
-        file_extensions = {'.ckpt', '.pt', '.pt2', '.bin', '.pth', '.safetensors', '.pkl', '.sft', '.gguf'}
+        file_extensions = {
+            ".ckpt",
+            ".pt",
+            ".pt2",
+            ".bin",
+            ".pth",
+            ".safetensors",
+            ".pkl",
+            ".sft",
+            ".gguf",
+        }
        super().__init__(
            model_type="checkpoint",
            model_class=CheckpointMetadata,
            file_extensions=file_extensions,
-            hash_index=ModelHashIndex()
+            hash_index=ModelHashIndex(),
        )

-    async def _create_default_metadata(self, file_path: str) -> Optional[CheckpointMetadata]:
+    async def _create_default_metadata(
+        self, file_path: str
+    ) -> Optional[CheckpointMetadata]:
        """Create default metadata for checkpoint without calculating hash (lazy hash).
-        
+
        Checkpoints are typically large (10GB+), so we skip hash calculation during initial
        scanning to improve startup performance. Hash will be calculated on-demand when
        fetching metadata from Civitai.
@@ -38,13 +51,13 @@ class CheckpointScanner(ModelScanner):
            if not os.path.exists(real_path):
                logger.error(f"File not found: {file_path}")
                return None
-            
+
            base_name = os.path.splitext(os.path.basename(file_path))[0]
            dir_path = os.path.dirname(file_path)
-            
+
            # Find preview image
            preview_url = find_preview_file(base_name, dir_path)
-            
+
            # Create metadata WITHOUT calculating hash
            metadata = CheckpointMetadata(
                file_name=base_name,
@@ -59,70 +72,76 @@ class CheckpointScanner(ModelScanner):
                modelDescription="",
                sub_type="checkpoint",
                from_civitai=False,  # Mark as local model since no hash yet
-                hash_status="pending"  # Mark hash as pending
+                hash_status="pending",  # Mark hash as pending
            )
-            
+
            # Save the created metadata
            logger.info(f"Creating checkpoint metadata (hash pending) for {file_path}")
            await MetadataManager.save_metadata(file_path, metadata)
-            
+
            return metadata
-            
+
        except Exception as e:
-            logger.error(f"Error creating default checkpoint metadata for {file_path}: {e}")
+            logger.error(
+                f"Error creating default checkpoint metadata for {file_path}: {e}"
+            )
            return None

    async def calculate_hash_for_model(self, file_path: str) -> Optional[str]:
        """Calculate hash for a checkpoint on-demand.
-        
+
        Args:
            file_path: Path to the model file
-            
+
        Returns:
            SHA256 hash string, or None if calculation failed
        """
        from ..utils.file_utils import calculate_sha256
-        
+
        try:
            real_path = os.path.realpath(file_path)
            if not os.path.exists(real_path):
                logger.error(f"File not found for hash calculation: {file_path}")
                return None
-            
+
            # Load current metadata
-            metadata, _ = await MetadataManager.load_metadata(file_path, self.model_class)
+            metadata, _ = await MetadataManager.load_metadata(
+                file_path, self.model_class
+            )
            if metadata is None:
                logger.error(f"No metadata found for {file_path}")
                return None
-            
+
            # Check if hash is already calculated
            if metadata.hash_status == "completed" and metadata.sha256:
                return metadata.sha256
-            
+
            # Update status to calculating
            metadata.hash_status = "calculating"
            await MetadataManager.save_metadata(file_path, metadata)
-            
+
            # Calculate hash
            logger.info(f"Calculating hash for checkpoint: {file_path}")
            sha256 = await calculate_sha256(real_path)
-            
+
            # Update metadata with hash
            metadata.sha256 = sha256
            metadata.hash_status = "completed"
            await MetadataManager.save_metadata(file_path, metadata)
-            
+
            # Update hash index
            self._hash_index.add_entry(sha256.lower(), file_path)
-            
+
            logger.info(f"Hash calculated for checkpoint: {file_path}")
            return sha256
-            
+
        except Exception as e:
            logger.error(f"Error calculating hash for {file_path}: {e}")
            # Update status to failed
            try:
-                metadata, _ = await MetadataManager.load_metadata(file_path, self.model_class)
+                metadata, _ = await MetadataManager.load_metadata(
+                    file_path, self.model_class
+                )
                if metadata:
                    metadata.hash_status = "failed"
                    await MetadataManager.save_metadata(file_path, metadata)
@@ -130,43 +149,46 @@ class CheckpointScanner(ModelScanner):
                pass
            return None

-    async def calculate_all_pending_hashes(self, progress_callback=None) -> Dict[str, int]:
+    async def calculate_all_pending_hashes(
+        self, progress_callback=None
+    ) -> Dict[str, int]:
        """Calculate hashes for all checkpoints with pending hash status.
-        
+
        If cache is not initialized, scans filesystem directly for metadata files
        with hash_status != 'completed'.
-        
+
        Args:
            progress_callback: Optional callback(progress, total, current_file)
-            
+
        Returns:
            Dict with 'completed', 'failed', 'total' counts
        """
        # Try to get from cache first
        cache = await self.get_cached_data()
-        
+
        if cache and cache.raw_data:
            # Use cache if available
            pending_models = [
-                item for item in cache.raw_data 
-                if item.get('hash_status') != 'completed' or not item.get('sha256')
+                item
+                for item in cache.raw_data
+                if item.get("hash_status") != "completed" or not item.get("sha256")
            ]
        else:
            # Cache not initialized, scan filesystem directly
            pending_models = await self._find_pending_models_from_filesystem()
-        
+
        if not pending_models:
-            return {'completed': 0, 'failed': 0, 'total': 0}
-        
+            return {"completed": 0, "failed": 0, "total": 0}
+
        total = len(pending_models)
        completed = 0
        failed = 0
-        
+
        for i, model_data in enumerate(pending_models):
-            file_path = model_data.get('file_path')
+            file_path = model_data.get("file_path")
            if not file_path:
                continue
-            
+
            try:
                sha256 = await self.calculate_hash_for_model(file_path)
                if sha256:
@@ -176,77 +198,102 @@ class CheckpointScanner(ModelScanner):
            except Exception as e:
                logger.error(f"Error calculating hash for {file_path}: {e}")
                failed += 1
-            
+
            if progress_callback:
                try:
                    await progress_callback(i + 1, total, file_path)
                except Exception:
                    pass
-        
-        return {
-            'completed': completed,
-            'failed': failed,
-            'total': total
-        }
-    
+
+        return {"completed": completed, "failed": failed, "total": total}
+
    async def _find_pending_models_from_filesystem(self) -> List[Dict[str, Any]]:
        """Scan filesystem for checkpoint metadata files with pending hash status."""
        pending_models = []
-        
+
        for root_path in self.get_model_roots():
            if not os.path.exists(root_path):
                continue
-            
+
            for dirpath, _dirnames, filenames in os.walk(root_path):
                for filename in filenames:
-                    if not filename.endswith('.metadata.json'):
+                    if not filename.endswith(".metadata.json"):
                        continue
-                    
+
                    metadata_path = os.path.join(dirpath, filename)
                    try:
-                        with open(metadata_path, 'r', encoding='utf-8') as f:
+                        with open(metadata_path, "r", encoding="utf-8") as f:
                            data = json.load(f)
-                        
+
                        # Check if hash is pending
-                        hash_status = data.get('hash_status', 'completed')
-                        sha256 = data.get('sha256', '')
-                        
-                        if hash_status != 'completed' or not sha256:
+                        hash_status = data.get("hash_status", "completed")
+                        sha256 = data.get("sha256", "")
+
+                        if hash_status != "completed" or not sha256:
                            # Find corresponding model file
-                            model_name = filename.replace('.metadata.json', '')
+                            model_name = filename.replace(".metadata.json", "")
                            model_path = None
-                            
+
                            # Look for model file with matching name
                            for ext in self.file_extensions:
                                potential_path = os.path.join(dirpath, model_name + ext)
                                if os.path.exists(potential_path):
                                    model_path = potential_path
                                    break
-                            
+
                            if model_path:
-                                pending_models.append({
-                                    'file_path': model_path.replace(os.sep, '/'),
-                                    'hash_status': hash_status,
-                                    'sha256': sha256,
-                                    **{k: v for k, v in data.items() if k not in ['file_path', 'hash_status', 'sha256']}
-                                })
+                                pending_models.append(
+                                    {
+                                        "file_path": model_path.replace(os.sep, "/"),
+                                        "hash_status": hash_status,
+                                        "sha256": sha256,
+                                        **{
+                                            k: v
+                                            for k, v in data.items()
+                                            if k
+                                            not in [
+                                                "file_path",
+                                                "hash_status",
+                                                "sha256",
+                                            ]
+                                        },
+                                    }
+                                )
                    except (json.JSONDecodeError, Exception) as e:
-                        logger.debug(f"Error reading metadata file {metadata_path}: {e}")
+                        logger.debug(
+                            f"Error reading metadata file {metadata_path}: {e}"
+                        )
                        continue
-        
+
        return pending_models

    def _resolve_sub_type(self, root_path: Optional[str]) -> Optional[str]:
-        """Resolve the sub-type based on the root path."""
+        """Resolve the sub-type based on the root path.
+
+        Checks both standard ComfyUI paths and LoRA Manager's extra folder paths.
+        """
        if not root_path:
            return None

+        # Check standard ComfyUI checkpoint paths
        if config.checkpoints_roots and root_path in config.checkpoints_roots:
            return "checkpoint"

+        # Check extra checkpoint paths
+        if (
+            config.extra_checkpoints_roots
+            and root_path in config.extra_checkpoints_roots
+        ):
+            return "checkpoint"
+
+        # Check standard ComfyUI unet paths
        if config.unet_roots and root_path in config.unet_roots:
            return "diffusion_model"

+        # Check extra unet paths
+        if config.extra_unet_roots and root_path in config.extra_unet_roots:
+            return "diffusion_model"
+
        return None

    def adjust_metadata(self, metadata, file_path, root_path):
--- a/py/services/custom_words_service.py
+++ b/py/services/custom_words_service.py
@@ -49,6 +49,7 @@ class CustomWordsService:
        if self._tag_index is None:
            try:
                from .tag_fts_index import get_tag_fts_index
+
                self._tag_index = get_tag_fts_index()
            except Exception as e:
                logger.warning(f"Failed to initialize TagFTSIndex: {e}")
@@ -59,14 +60,16 @@ class CustomWordsService:
        self,
        search_term: str,
        limit: int = 20,
+        offset: int = 0,
        categories: Optional[List[int]] = None,
-        enriched: bool = False
+        enriched: bool = False,
    ) -> List[Dict[str, Any]]:
        """Search tags using TagFTSIndex with category filtering.

        Args:
            search_term: The search term to match against.
            limit: Maximum number of results to return.
+            offset: Number of results to skip.
            categories: Optional list of category IDs to filter by.
            enriched: If True, always return enriched results with category
                       and post_count (default behavior now).
@@ -76,7 +79,9 @@ class CustomWordsService:
        """
        tag_index = self._get_tag_index()
        if tag_index is not None:
-            results = tag_index.search(search_term, categories=categories, limit=limit)
+            results = tag_index.search(
+                search_term, categories=categories, limit=limit, offset=offset
+            )
            return results

        logger.debug("TagFTSIndex not available, returning empty results")
--- a/py/services/download_manager.py
+++ b/py/services/download_manager.py
@@ -10,7 +10,11 @@ import uuid
 from typing import Dict, List, Optional, Set, Tuple
 from urllib.parse import urlparse
 from ..utils.models import LoraMetadata, CheckpointMetadata, EmbeddingMetadata
-from ..utils.constants import CARD_PREVIEW_WIDTH, DIFFUSION_MODEL_BASE_MODELS, VALID_LORA_TYPES
+from ..utils.constants import (
+    CARD_PREVIEW_WIDTH,
+    DIFFUSION_MODEL_BASE_MODELS,
+    VALID_LORA_TYPES,
+)
 from ..utils.civitai_utils import rewrite_preview_url
 from ..utils.preview_selection import select_preview_media
 from ..utils.utils import sanitize_folder_name
@@ -352,10 +356,12 @@ class DownloadManager:
            # Check if this checkpoint should be treated as a diffusion model based on baseModel
            is_diffusion_model = False
            if model_type == "checkpoint":
-                base_model_value = version_info.get('baseModel', '')
+                base_model_value = version_info.get("baseModel", "")
                if base_model_value in DIFFUSION_MODEL_BASE_MODELS:
                    is_diffusion_model = True
-                    logger.info(f"baseModel '{base_model_value}' is a known diffusion model, routing to unet folder")
+                    logger.info(
+                        f"baseModel '{base_model_value}' is a known diffusion model, routing to unet folder"
+                    )

            # Case 2: model_version_id was None, check after getting version_info
            if model_version_id is None:
@@ -464,7 +470,7 @@ class DownloadManager:
            # 2. Get file information
            files = version_info.get("files", [])
            file_info = None
-            
+
            # If file_params is provided, try to find matching file
            if file_params and model_version_id:
                target_type = file_params.get("type", "Model")
@@ -472,23 +478,28 @@ class DownloadManager:
                target_size = file_params.get("size", "full")
                target_fp = file_params.get("fp")
                is_primary = file_params.get("isPrimary", False)
-                
+
                if is_primary:
                    # Find primary file
                    file_info = next(
-                        (f for f in files if f.get("primary") and f.get("type") in ("Model", "Negative")),
-                        None
+                        (
+                            f
+                            for f in files
+                            if f.get("primary")
+                            and f.get("type") in ("Model", "Negative")
+                        ),
+                        None,
                    )
                else:
                    # Match by metadata
                    for f in files:
                        f_type = f.get("type", "")
                        f_meta = f.get("metadata", {})
-                        
+
                        # Check type match
                        if f_type != target_type:
                            continue
-                        
+
                        # Check metadata match
                        if f_meta.get("format") != target_format:
                            continue
@@ -496,10 +507,10 @@ class DownloadManager:
                            continue
                        if target_fp and f_meta.get("fp") != target_fp:
                            continue
-                        
+
                        file_info = f
                        break
-            
+
            # Fallback to primary file if no match found
            if not file_info:
                file_info = next(
@@ -510,7 +521,7 @@ class DownloadManager:
                    ),
                    None,
                )
-            
+
            if not file_info:
                return {"success": False, "error": "No suitable file found in metadata"}
            mirrors = file_info.get("mirrors") or []
@@ -1220,7 +1231,13 @@ class DownloadManager:
        entries: List = []
        for index, file_path in enumerate(file_paths):
            entry = base_metadata if index == 0 else copy.deepcopy(base_metadata)
-            entry.update_file_info(file_path)
+            # Update file paths without modifying size and modified timestamps
+            # modified should remain as the download start time (import time)
+            # size will be updated below to reflect actual downloaded file size
+            entry.file_path = file_path.replace(os.sep, "/")
+            entry.file_name = os.path.splitext(os.path.basename(file_path))[0]
+            # Update size to actual downloaded file size
+            entry.size = os.path.getsize(file_path)
            entry.sha256 = await calculate_sha256(file_path)
            entries.append(entry)

--- a/py/services/lora_service.py
+++ b/py/services/lora_service.py
@@ -516,12 +516,18 @@ class LoraService(BaseModelService):
        if sort_by == "model_name":
            available_loras = sorted(
                available_loras,
-                key=lambda x: (x.get("model_name") or x.get("file_name", "")).lower()
+                key=lambda x: (
+                    (x.get("model_name") or x.get("file_name", "")).lower(),
+                    x.get("file_path", "").lower()
+                )
            )
        else:  # Default to filename
            available_loras = sorted(
                available_loras,
-                key=lambda x: x.get("file_name", "").lower()
+                key=lambda x: (
+                    x.get("file_name", "").lower(),
+                    x.get("file_path", "").lower()
+                )
            )

        # Return minimal data needed for cycling
--- a/py/services/model_cache.py
+++ b/py/services/model_cache.py
@@ -221,33 +221,45 @@ class ModelCache:
        start_time = time.perf_counter()
        reverse = (order == 'desc')
        if sort_key == 'name':
-            # Natural sort by configured display name, case-insensitive
+            # Natural sort by configured display name, case-insensitive, with file_path as tie-breaker
            result = natsorted(
                data,
-                key=lambda x: self._get_display_name(x).lower(),
+                key=lambda x: (
+                    self._get_display_name(x).lower(),
+                    x.get('file_path', '').lower()
+                ),
                reverse=reverse
            )
        elif sort_key == 'date':
-            # Sort by modified timestamp (use .get() with default to handle missing fields)
+            # Sort by modified timestamp, fallback to name and path for stability
            result = sorted(
                data,
-                key=lambda x: x.get('modified', 0.0),
+                key=lambda x: (
+                    x.get('modified', 0.0),
+                    self._get_display_name(x).lower(),
+                    x.get('file_path', '').lower()
+                ),
                reverse=reverse
            )
        elif sort_key == 'size':
-            # Sort by file size (use .get() with default to handle missing fields)
+            # Sort by file size, fallback to name and path for stability
            result = sorted(
                data,
-                key=lambda x: x.get('size', 0),
+                key=lambda x: (
+                    x.get('size', 0),
+                    self._get_display_name(x).lower(),
+                    x.get('file_path', '').lower()
+                ),
                reverse=reverse
            )
        elif sort_key == 'usage':
-            # Sort by usage count, fallback to 0, then name for stability
+            # Sort by usage count, fallback to 0, then name and path for stability
            return sorted(
                data,
                key=lambda x: (
                    x.get('usage_count', 0),
-                    self._get_display_name(x).lower()
+                    self._get_display_name(x).lower(),
+                    x.get('file_path', '').lower()
                ),
                reverse=reverse
            )
--- a/py/services/recipe_cache.py
+++ b/py/services/recipe_cache.py
@@ -4,6 +4,7 @@ from dataclasses import dataclass
 from operator import itemgetter
 from natsort import natsorted

+
@dataclass
 class RecipeCache:
    """Cache structure for Recipe data"""
@@ -21,11 +22,18 @@ class RecipeCache:
        self.folder_tree = self.folder_tree or {}

    async def resort(self, name_only: bool = False):
-        """Resort all cached data views"""
+        """Resort all cached data views in a thread pool to avoid blocking the event loop."""
        async with self._lock:
-            self._resort_locked(name_only=name_only)
+            loop = asyncio.get_event_loop()
+            await loop.run_in_executor(
+                None,
+                self._resort_locked,
+                name_only,
+            )

-    async def update_recipe_metadata(self, recipe_id: str, metadata: Dict, *, resort: bool = True) -> bool:
+    async def update_recipe_metadata(
+        self, recipe_id: str, metadata: Dict, *, resort: bool = True
+    ) -> bool:
        """Update metadata for a specific recipe in all cached data

        Args:
@@ -37,7 +45,7 @@ class RecipeCache:
        """
        async with self._lock:
            for item in self.raw_data:
-                if str(item.get('id')) == str(recipe_id):
+                if str(item.get("id")) == str(recipe_id):
                    item.update(metadata)
                    if resort:
                        self._resort_locked()
@@ -52,7 +60,9 @@ class RecipeCache:
            if resort:
                self._resort_locked()

-    async def remove_recipe(self, recipe_id: str, *, resort: bool = False) -> Optional[Dict]:
+    async def remove_recipe(
+        self, recipe_id: str, *, resort: bool = False
+    ) -> Optional[Dict]:
        """Remove a recipe from the cache by ID.

        Args:
@@ -64,14 +74,16 @@ class RecipeCache:

        async with self._lock:
            for index, recipe in enumerate(self.raw_data):
-                if str(recipe.get('id')) == str(recipe_id):
+                if str(recipe.get("id")) == str(recipe_id):
                    removed = self.raw_data.pop(index)
                    if resort:
                        self._resort_locked()
                    return removed
        return None

-    async def bulk_remove(self, recipe_ids: Iterable[str], *, resort: bool = False) -> List[Dict]:
+    async def bulk_remove(
+        self, recipe_ids: Iterable[str], *, resort: bool = False
+    ) -> List[Dict]:
        """Remove multiple recipes from the cache."""

        id_set = {str(recipe_id) for recipe_id in recipe_ids}
@@ -79,21 +91,25 @@ class RecipeCache:
            return []

        async with self._lock:
-            removed = [item for item in self.raw_data if str(item.get('id')) in id_set]
+            removed = [item for item in self.raw_data if str(item.get("id")) in id_set]
            if not removed:
                return []

-            self.raw_data = [item for item in self.raw_data if str(item.get('id')) not in id_set]
+            self.raw_data = [
+                item for item in self.raw_data if str(item.get("id")) not in id_set
+            ]
            if resort:
                self._resort_locked()
            return removed

-    async def replace_recipe(self, recipe_id: str, new_data: Dict, *, resort: bool = False) -> bool:
+    async def replace_recipe(
+        self, recipe_id: str, new_data: Dict, *, resort: bool = False
+    ) -> bool:
        """Replace cached data for a recipe."""

        async with self._lock:
            for index, recipe in enumerate(self.raw_data):
-                if str(recipe.get('id')) == str(recipe_id):
+                if str(recipe.get("id")) == str(recipe_id):
                    self.raw_data[index] = new_data
                    if resort:
                        self._resort_locked()
@@ -105,7 +121,7 @@ class RecipeCache:

        async with self._lock:
            for recipe in self.raw_data:
-                if str(recipe.get('id')) == str(recipe_id):
+                if str(recipe.get("id")) == str(recipe_id):
                    return dict(recipe)
        return None

@@ -115,16 +131,14 @@ class RecipeCache:
        async with self._lock:
            return [dict(item) for item in self.raw_data]

-    def _resort_locked(self, *, name_only: bool = False) -> None:
+    def _resort_locked(self, name_only: bool = False) -> None:
        """Sort cached views. Caller must hold ``_lock``."""

        self.sorted_by_name = natsorted(
            self.raw_data,
-            key=lambda x: x.get('title', '').lower()
+            key=lambda x: (x.get("title", "").lower(), x.get("file_path", "").lower()),
        )
        if not name_only:
            self.sorted_by_date = sorted(
-                self.raw_data,
-                key=itemgetter('created_date', 'file_path'),
-                reverse=True
-            )
+                self.raw_data, key=itemgetter("created_date", "file_path"), reverse=True
+            )
--- a/py/services/recipe_scanner.py
+++ b/py/services/recipe_scanner.py
--- a/py/services/tag_fts_index.py
+++ b/py/services/tag_fts_index.py
@@ -69,7 +69,9 @@ class TagFTSIndex:
    _DEFAULT_FILENAME = "tag_fts.sqlite"
    _CSV_FILENAME = "danbooru_e621_merged.csv"

-    def __init__(self, db_path: Optional[str] = None, csv_path: Optional[str] = None) -> None:
+    def __init__(
+        self, db_path: Optional[str] = None, csv_path: Optional[str] = None
+    ) -> None:
        """Initialize the FTS index.

        Args:
@@ -92,7 +94,9 @@ class TagFTSIndex:
            if directory:
                os.makedirs(directory, exist_ok=True)
        except Exception as exc:
-            logger.warning("Could not create FTS index directory %s: %s", directory, exc)
+            logger.warning(
+                "Could not create FTS index directory %s: %s", directory, exc
+            )

    def _resolve_default_db_path(self) -> str:
        """Resolve the default database path."""
@@ -173,13 +177,15 @@ class TagFTSIndex:
                    # Set schema version
                    conn.execute(
                        "INSERT OR REPLACE INTO fts_metadata (key, value) VALUES (?, ?)",
-                        ("schema_version", str(SCHEMA_VERSION))
+                        ("schema_version", str(SCHEMA_VERSION)),
                    )
                    conn.commit()

                    self._schema_initialized = True
                    self._needs_rebuild = needs_rebuild
-                    logger.debug("Tag FTS index schema initialized at %s", self._db_path)
+                    logger.debug(
+                        "Tag FTS index schema initialized at %s", self._db_path
+                    )
                finally:
                    conn.close()
            except Exception as exc:
@@ -206,13 +212,20 @@ class TagFTSIndex:
            row = cursor.fetchone()
            if not row:
                # Old schema without version, needs rebuild
-                logger.info("Migrating tag FTS index to schema version %d (adding alias support)", SCHEMA_VERSION)
+                logger.info(
+                    "Migrating tag FTS index to schema version %d (adding alias support)",
+                    SCHEMA_VERSION,
+                )
                self._drop_old_tables(conn)
                return True

            current_version = int(row[0])
            if current_version < SCHEMA_VERSION:
-                logger.info("Migrating tag FTS index from version %d to %d", current_version, SCHEMA_VERSION)
+                logger.info(
+                    "Migrating tag FTS index from version %d to %d",
+                    current_version,
+                    SCHEMA_VERSION,
+                )
                self._drop_old_tables(conn)
                return True

@@ -246,7 +259,9 @@ class TagFTSIndex:
            return

        if not os.path.exists(self._csv_path):
-            logger.warning("CSV file not found at %s, cannot build tag index", self._csv_path)
+            logger.warning(
+                "CSV file not found at %s, cannot build tag index", self._csv_path
+            )
            return

        self._indexing_in_progress = True
@@ -314,22 +329,24 @@ class TagFTSIndex:
                    # Update metadata
                    conn.execute(
                        "INSERT OR REPLACE INTO fts_metadata (key, value) VALUES (?, ?)",
-                        ("last_build_time", str(time.time()))
+                        ("last_build_time", str(time.time())),
                    )
                    conn.execute(
                        "INSERT OR REPLACE INTO fts_metadata (key, value) VALUES (?, ?)",
-                        ("tag_count", str(total_inserted))
+                        ("tag_count", str(total_inserted)),
                    )
                    conn.execute(
                        "INSERT OR REPLACE INTO fts_metadata (key, value) VALUES (?, ?)",
-                        ("schema_version", str(SCHEMA_VERSION))
+                        ("schema_version", str(SCHEMA_VERSION)),
                    )

                    conn.commit()
                    elapsed = time.time() - start_time
                    logger.info(
                        "Tag FTS index built: %d tags indexed (%d with aliases) in %.2fs",
-                        total_inserted, tags_with_aliases, elapsed
+                        total_inserted,
+                        tags_with_aliases,
+                        elapsed,
                    )
                finally:
                    conn.close()
@@ -350,7 +367,7 @@ class TagFTSIndex:
        # Insert into tags table (with aliases)
        conn.executemany(
            "INSERT OR IGNORE INTO tags (tag_name, category, post_count, aliases) VALUES (?, ?, ?, ?)",
-            rows
+            rows,
        )

        # Build a map of tag_name -> aliases for FTS insertion
@@ -362,7 +379,7 @@ class TagFTSIndex:
        placeholders = ",".join("?" * len(tag_names))
        cursor = conn.execute(
            f"SELECT rowid, tag_name FROM tags WHERE tag_name IN ({placeholders})",
-            tag_names
+            tag_names,
        )

        # Build FTS rows with (rowid, searchable_text) = (tags.rowid, "tag_name alias1 alias2 ...")
@@ -379,13 +396,17 @@ class TagFTSIndex:
                        alias = alias[1:]  # Remove leading slash
                    if alias:
                        alias_parts.append(alias)
-                searchable_text = f"{tag_name} {' '.join(alias_parts)}" if alias_parts else tag_name
+                searchable_text = (
+                    f"{tag_name} {' '.join(alias_parts)}" if alias_parts else tag_name
+                )
            else:
                searchable_text = tag_name
            fts_rows.append((rowid, searchable_text))

        if fts_rows:
-            conn.executemany("INSERT INTO tag_fts (rowid, searchable_text) VALUES (?, ?)", fts_rows)
+            conn.executemany(
+                "INSERT INTO tag_fts (rowid, searchable_text) VALUES (?, ?)", fts_rows
+            )

    def ensure_ready(self) -> bool:
        """Ensure the index is ready, building if necessary.
@@ -420,21 +441,28 @@ class TagFTSIndex:
        self,
        query: str,
        categories: Optional[List[int]] = None,
-        limit: int = 20
+        limit: int = 20,
+        offset: int = 0,
    ) -> List[Dict]:
        """Search tags using FTS5 with prefix matching.

        Supports alias search: if the query matches an alias rather than
        the tag_name, the result will include a "matched_alias" field.

+        Ranking is based on a combination of:
+        1. FTS5 bm25 relevance score (how well the text matches)
+        2. Post count (popularity)
+        3. Exact prefix match boost (tag_name starts with query)
+
        Args:
            query: The search query string.
            categories: Optional list of category IDs to filter by.
            limit: Maximum number of results to return.
+            offset: Number of results to skip.

        Returns:
            List of dictionaries with tag_name, category, post_count,
-            and optionally matched_alias.
+            rank_score, and optionally matched_alias.
        """
        # Ensure index is ready (lazy initialization)
        if not self.ensure_ready():
@@ -450,35 +478,67 @@ class TagFTSIndex:
        if not fts_query:
            return []

+        query_lower = query.lower().strip()
+
        try:
            with self._lock:
                conn = self._connect(readonly=True)
                try:
-                    # Build the SQL query - now also fetch aliases for matched_alias detection
-                    # Use subquery for category filter to ensure FTS is evaluated first
+                    # Build the SQL query with bm25 ranking
+                    # FTS5 bm25() returns negative scores, lower is better
+                    # We use -bm25() to get higher=better scores
+                    # Weights: -100.0 for exact matches, 1.0 for others
+                    # Add LOG10(post_count) weighting to boost popular tags
+                    # Use CASE to boost tag_name prefix matches above alias matches
                    if categories:
                        placeholders = ",".join("?" * len(categories))
                        sql = f"""
-                            SELECT t.tag_name, t.category, t.post_count, t.aliases
-                            FROM tags t
-                            WHERE t.rowid IN (
-                                SELECT rowid FROM tag_fts WHERE searchable_text MATCH ?
-                            )
+                            SELECT t.tag_name, t.category, t.post_count, t.aliases,
+                                   CASE 
+                                       WHEN t.tag_name LIKE ? ESCAPE '\\' THEN 1
+                                       ELSE 0
+                                   END AS is_tag_name_match,
+                                   bm25(tag_fts, -100.0, 1.0, 1.0) + LOG10(t.post_count + 1) * 10.0 AS rank_score
+                            FROM tag_fts
+                            JOIN tags t ON tag_fts.rowid = t.rowid
+                            WHERE tag_fts.searchable_text MATCH ?
                            AND t.category IN ({placeholders})
-                            ORDER BY t.post_count DESC
-                            LIMIT ?
+                            ORDER BY is_tag_name_match DESC, rank_score DESC
+                            LIMIT ? OFFSET ?
                        """
-                        params = [fts_query] + categories + [limit]
+                        # Escape special LIKE characters and add wildcard
+                        query_escaped = (
+                            query_lower.lstrip("/")
+                            .replace("\\", "\\\\")
+                            .replace("%", "\\%")
+                            .replace("_", "\\_")
+                        )
+                        params = (
+                            [query_escaped + "%", fts_query]
+                            + categories
+                            + [limit, offset]
+                        )
                    else:
                        sql = """
-                            SELECT t.tag_name, t.category, t.post_count, t.aliases
-                            FROM tag_fts f
-                            JOIN tags t ON f.rowid = t.rowid
-                            WHERE f.searchable_text MATCH ?
-                            ORDER BY t.post_count DESC
-                            LIMIT ?
+                            SELECT t.tag_name, t.category, t.post_count, t.aliases,
+                                   CASE 
+                                       WHEN t.tag_name LIKE ? ESCAPE '\\' THEN 1
+                                       ELSE 0
+                                   END AS is_tag_name_match,
+                                   bm25(tag_fts, -100.0, 1.0, 1.0) + LOG10(t.post_count + 1) * 10.0 AS rank_score
+                            FROM tag_fts
+                            JOIN tags t ON tag_fts.rowid = t.rowid
+                            WHERE tag_fts.searchable_text MATCH ?
+                            ORDER BY is_tag_name_match DESC, rank_score DESC
+                            LIMIT ? OFFSET ?
                        """
-                        params = [fts_query, limit]
+                        query_escaped = (
+                            query_lower.lstrip("/")
+                            .replace("\\", "\\\\")
+                            .replace("%", "\\%")
+                            .replace("_", "\\_")
+                        )
+                        params = [query_escaped + "%", fts_query, limit, offset]

                    cursor = conn.execute(sql, params)
                    results = []
@@ -487,8 +547,17 @@ class TagFTSIndex:
                            "tag_name": row[0],
                            "category": row[1],
                            "post_count": row[2],
+                            "is_tag_name_match": row[4] == 1,
+                            "rank_score": row[5],
                        }

+                        # Set is_exact_prefix based on tag_name match
+                        tag_name = row[0]
+                        if tag_name.lower().startswith(query_lower.lstrip("/")):
+                            result["is_exact_prefix"] = True
+                        else:
+                            result["is_exact_prefix"] = result["is_tag_name_match"]
+
                        # Check if search matched an alias rather than the tag_name
                        matched_alias = self._find_matched_alias(query, row[0], row[3])
                        if matched_alias:
@@ -502,7 +571,9 @@ class TagFTSIndex:
            logger.debug("Tag FTS search error for query '%s': %s", query, exc)
            return []

-    def _find_matched_alias(self, query: str, tag_name: str, aliases_str: str) -> Optional[str]:
+    def _find_matched_alias(
+        self, query: str, tag_name: str, aliases_str: str
+    ) -> Optional[str]:
        """Find which alias matched the query, if any.

        Args:
--- a/py/utils/models.py
+++ b/py/utils/models.py
@@ -4,32 +4,40 @@ from datetime import datetime
 import os
 from .model_utils import determine_base_model

+
@dataclass
 class BaseModelMetadata:
    """Base class for all model metadata structures"""
-    file_name: str              # The filename without extension
-    model_name: str             # The model's name defined by the creator
-    file_path: str              # Full path to the model file
-    size: int                   # File size in bytes
-    modified: float             # Timestamp when the model was added to the management system
-    sha256: str                 # SHA256 hash of the file
-    base_model: str             # Base model type (SD1.5/SD2.1/SDXL/etc.)
-    preview_url: str            # Preview image URL
-    preview_nsfw_level: int = 0 # NSFW level of the preview image
-    notes: str = ""             # Additional notes
-    from_civitai: bool = True   # Whether from Civitai
-    civitai: Dict[str, Any] = field(default_factory=dict)  # Civitai API data if available
-    tags: List[str] = None      # Model tags
+
+    file_name: str  # The filename without extension
+    model_name: str  # The model's name defined by the creator
+    file_path: str  # Full path to the model file
+    size: int  # File size in bytes
+    modified: float  # Timestamp when the model was added to the management system
+    sha256: str  # SHA256 hash of the file
+    base_model: str  # Base model type (SD1.5/SD2.1/SDXL/etc.)
+    preview_url: str  # Preview image URL
+    preview_nsfw_level: int = 0  # NSFW level of the preview image
+    notes: str = ""  # Additional notes
+    from_civitai: bool = True  # Whether from Civitai
+    civitai: Dict[str, Any] = field(
+        default_factory=dict
+    )  # Civitai API data if available
+    tags: List[str] = None  # Model tags
    modelDescription: str = ""  # Full model description
    civitai_deleted: bool = False  # Whether deleted from Civitai
-    favorite: bool = False      # Whether the model is a favorite
-    exclude: bool = False       # Whether to exclude this model from the cache
-    db_checked: bool = False    # Whether checked in archive DB
-    skip_metadata_refresh: bool = False  # Whether to skip this model during bulk metadata refresh
+    favorite: bool = False  # Whether the model is a favorite
+    exclude: bool = False  # Whether to exclude this model from the cache
+    db_checked: bool = False  # Whether checked in archive DB
+    skip_metadata_refresh: bool = (
+        False  # Whether to skip this model during bulk metadata refresh
+    )
    metadata_source: Optional[str] = None  # Last provider that supplied metadata
    last_checked_at: float = 0  # Last checked timestamp
    hash_status: str = "completed"  # Hash calculation status: pending | calculating | completed | failed
-    _unknown_fields: Dict[str, Any] = field(default_factory=dict, repr=False, compare=False)  # Store unknown fields
+    _unknown_fields: Dict[str, Any] = field(
+        default_factory=dict, repr=False, compare=False
+    )  # Store unknown fields

    def __post_init__(self):
        # Initialize empty lists to avoid mutable default parameter issue
@@ -40,211 +48,238 @@ class BaseModelMetadata:
            self.tags = []

    @classmethod
-    def from_dict(cls, data: Dict) -> 'BaseModelMetadata':
+    def from_dict(cls, data: Dict) -> "BaseModelMetadata":
        """Create instance from dictionary"""
        data_copy = data.copy()
-        
+
        # Use cached fields if available, otherwise compute them
-        if not hasattr(cls, '_known_fields_cache'):
+        if not hasattr(cls, "_known_fields_cache"):
            known_fields = set()
            for c in cls.mro():
-                if hasattr(c, '__annotations__'):
+                if hasattr(c, "__annotations__"):
                    known_fields.update(c.__annotations__.keys())
            cls._known_fields_cache = known_fields
-        
+
        known_fields = cls._known_fields_cache
-        
+
        # Extract fields that match our class attributes
        fields_to_use = {k: v for k, v in data_copy.items() if k in known_fields}
-        
+
        # Store unknown fields separately
-        unknown_fields = {k: v for k, v in data_copy.items() if k not in known_fields and not k.startswith('_')}
-        
+        unknown_fields = {
+            k: v
+            for k, v in data_copy.items()
+            if k not in known_fields and not k.startswith("_")
+        }
+
        # Create instance with known fields
        instance = cls(**fields_to_use)
-        
+
        # Add unknown fields as a separate attribute
        instance._unknown_fields = unknown_fields
-        
+
        return instance

    def to_dict(self) -> Dict:
        """Convert to dictionary for JSON serialization"""
        result = asdict(self)
-        
+
        # Remove private fields
-        result = {k: v for k, v in result.items() if not k.startswith('_')}
-        
+        result = {k: v for k, v in result.items() if not k.startswith("_")}
+
        # Add back unknown fields if they exist
-        if hasattr(self, '_unknown_fields'):
+        if hasattr(self, "_unknown_fields"):
            result.update(self._unknown_fields)
-            
+
        return result

    def update_civitai_info(self, civitai_data: Dict) -> None:
        """Update Civitai information"""
        self.civitai = civitai_data

-    def update_file_info(self, file_path: str) -> None:
-        """Update metadata with actual file information"""
+    def update_file_info(self, file_path: str, update_timestamps: bool = False) -> None:
+        """
+        Update metadata with actual file information.
+
+        Args:
+            file_path: Path to the model file
+            update_timestamps: If True, update size and modified from filesystem.
+                              If False (default), only update file_path and file_name.
+                              Set to True only when file has been moved/relocated.
+        """
        if os.path.exists(file_path):
-            self.size = os.path.getsize(file_path)
-            self.modified = os.path.getmtime(file_path)
-            self.file_path = file_path.replace(os.sep, '/')
-            # Update file_name when file_path changes
+            if update_timestamps:
+                # Only update size and modified when file has been relocated
+                self.size = os.path.getsize(file_path)
+                self.modified = os.path.getmtime(file_path)
+            # Always update paths when this method is called
+            self.file_path = file_path.replace(os.sep, "/")
            self.file_name = os.path.splitext(os.path.basename(file_path))[0]

    @staticmethod
-    def generate_unique_filename(target_dir: str, base_name: str, extension: str, hash_provider: callable = None) -> str:
+    def generate_unique_filename(
+        target_dir: str, base_name: str, extension: str, hash_provider: callable = None
+    ) -> str:
        """Generate a unique filename to avoid conflicts
-        
+
        Args:
            target_dir: Target directory path
            base_name: Base filename without extension
            extension: File extension including the dot
            hash_provider: A callable that returns the SHA256 hash when needed
-            
+
        Returns:
            str: Unique filename that doesn't conflict with existing files
        """
        original_filename = f"{base_name}{extension}"
        target_path = os.path.join(target_dir, original_filename)
-        
+
        # If no conflict, return original filename
        if not os.path.exists(target_path):
            return original_filename
-            
+
        # Only compute hash when needed
        if hash_provider:
            sha256_hash = hash_provider()
        else:
            sha256_hash = "0000"
-        
+
        # Generate short hash (first 4 characters of SHA256)
        short_hash = sha256_hash[:4] if sha256_hash else "0000"
-        
+
        # Try with short hash suffix
        unique_filename = f"{base_name}-{short_hash}{extension}"
        unique_path = os.path.join(target_dir, unique_filename)
-        
+
        # If still conflicts, add incremental number
        counter = 1
        while os.path.exists(unique_path):
            unique_filename = f"{base_name}-{short_hash}-{counter}{extension}"
            unique_path = os.path.join(target_dir, unique_filename)
            counter += 1
-            
+
        return unique_filename

+
@dataclass
 class LoraMetadata(BaseModelMetadata):
    """Represents the metadata structure for a Lora model"""
-    usage_tips: str = "{}"      # Usage tips for the model, json string
+
+    usage_tips: str = "{}"  # Usage tips for the model, json string

    @classmethod
-    def from_civitai_info(cls, version_info: Dict, file_info: Dict, save_path: str) -> 'LoraMetadata':
+    def from_civitai_info(
+        cls, version_info: Dict, file_info: Dict, save_path: str
+    ) -> "LoraMetadata":
        """Create LoraMetadata instance from Civitai version info"""
-        file_name = file_info.get('name', '')
-        base_model = determine_base_model(version_info.get('baseModel', ''))
+        file_name = file_info.get("name", "")
+        base_model = determine_base_model(version_info.get("baseModel", ""))

        # Extract tags and description if available
        tags = []
        description = ""
-        model_data = version_info.get('model') or {}
-        if 'tags' in model_data:
-            tags = model_data['tags']
-        if 'description' in model_data:
-            description = model_data['description']
+        model_data = version_info.get("model") or {}
+        if "tags" in model_data:
+            tags = model_data["tags"]
+        if "description" in model_data:
+            description = model_data["description"]

        return cls(
            file_name=os.path.splitext(file_name)[0],
-            model_name=model_data.get('name', os.path.splitext(file_name)[0]),
-            file_path=save_path.replace(os.sep, '/'),
-            size=file_info.get('sizeKB', 0) * 1024,
+            model_name=model_data.get("name", os.path.splitext(file_name)[0]),
+            file_path=save_path.replace(os.sep, "/"),
+            size=file_info.get("sizeKB", 0) * 1024,
            modified=datetime.now().timestamp(),
-            sha256=(file_info.get('hashes') or {}).get('SHA256', '').lower(),
+            sha256=(file_info.get("hashes") or {}).get("SHA256", "").lower(),
            base_model=base_model,
-            preview_url='',  # Will be updated after preview download
-            preview_nsfw_level=0, # Will be updated after preview download
+            preview_url="",  # Will be updated after preview download
+            preview_nsfw_level=0,  # Will be updated after preview download
            from_civitai=True,
            civitai=version_info,
            tags=tags,
-            modelDescription=description
+            modelDescription=description,
        )

+
@dataclass
 class CheckpointMetadata(BaseModelMetadata):
    """Represents the metadata structure for a Checkpoint model"""
+
    sub_type: str = "checkpoint"  # Model sub-type (checkpoint, diffusion_model, etc.)

    @classmethod
-    def from_civitai_info(cls, version_info: Dict, file_info: Dict, save_path: str) -> 'CheckpointMetadata':
+    def from_civitai_info(
+        cls, version_info: Dict, file_info: Dict, save_path: str
+    ) -> "CheckpointMetadata":
        """Create CheckpointMetadata instance from Civitai version info"""
-        file_name = file_info.get('name', '')
-        base_model = determine_base_model(version_info.get('baseModel', ''))
-        sub_type = version_info.get('type', 'checkpoint')
+        file_name = file_info.get("name", "")
+        base_model = determine_base_model(version_info.get("baseModel", ""))
+        sub_type = version_info.get("type", "checkpoint")

        # Extract tags and description if available
        tags = []
        description = ""
-        model_data = version_info.get('model') or {}
-        if 'tags' in model_data:
-            tags = model_data['tags']
-        if 'description' in model_data:
-            description = model_data['description']
+        model_data = version_info.get("model") or {}
+        if "tags" in model_data:
+            tags = model_data["tags"]
+        if "description" in model_data:
+            description = model_data["description"]

        return cls(
            file_name=os.path.splitext(file_name)[0],
-            model_name=model_data.get('name', os.path.splitext(file_name)[0]),
-            file_path=save_path.replace(os.sep, '/'),
-            size=file_info.get('sizeKB', 0) * 1024,
+            model_name=model_data.get("name", os.path.splitext(file_name)[0]),
+            file_path=save_path.replace(os.sep, "/"),
+            size=file_info.get("sizeKB", 0) * 1024,
            modified=datetime.now().timestamp(),
-            sha256=(file_info.get('hashes') or {}).get('SHA256', '').lower(),
+            sha256=(file_info.get("hashes") or {}).get("SHA256", "").lower(),
            base_model=base_model,
-            preview_url='',  # Will be updated after preview download
+            preview_url="",  # Will be updated after preview download
            preview_nsfw_level=0,
            from_civitai=True,
            civitai=version_info,
            sub_type=sub_type,
            tags=tags,
-            modelDescription=description
+            modelDescription=description,
        )

+
@dataclass
 class EmbeddingMetadata(BaseModelMetadata):
    """Represents the metadata structure for an Embedding model"""
+
    sub_type: str = "embedding"

    @classmethod
-    def from_civitai_info(cls, version_info: Dict, file_info: Dict, save_path: str) -> 'EmbeddingMetadata':
+    def from_civitai_info(
+        cls, version_info: Dict, file_info: Dict, save_path: str
+    ) -> "EmbeddingMetadata":
        """Create EmbeddingMetadata instance from Civitai version info"""
-        file_name = file_info.get('name', '')
-        base_model = determine_base_model(version_info.get('baseModel', ''))
-        sub_type = version_info.get('type', 'embedding')
+        file_name = file_info.get("name", "")
+        base_model = determine_base_model(version_info.get("baseModel", ""))
+        sub_type = version_info.get("type", "embedding")

        # Extract tags and description if available
        tags = []
        description = ""
-        model_data = version_info.get('model') or {}
-        if 'tags' in model_data:
-            tags = model_data['tags']
-        if 'description' in model_data:
-            description = model_data['description']
+        model_data = version_info.get("model") or {}
+        if "tags" in model_data:
+            tags = model_data["tags"]
+        if "description" in model_data:
+            description = model_data["description"]

        return cls(
            file_name=os.path.splitext(file_name)[0],
-            model_name=model_data.get('name', os.path.splitext(file_name)[0]),
-            file_path=save_path.replace(os.sep, '/'),
-            size=file_info.get('sizeKB', 0) * 1024,
+            model_name=model_data.get("name", os.path.splitext(file_name)[0]),
+            file_path=save_path.replace(os.sep, "/"),
+            size=file_info.get("sizeKB", 0) * 1024,
            modified=datetime.now().timestamp(),
-            sha256=(file_info.get('hashes') or {}).get('SHA256', '').lower(),
+            sha256=(file_info.get("hashes") or {}).get("SHA256", "").lower(),
            base_model=base_model,
-            preview_url='',  # Will be updated after preview download
+            preview_url="",  # Will be updated after preview download
            preview_nsfw_level=0,
            from_civitai=True,
            civitai=version_info,
            sub_type=sub_type,
            tags=tags,
-            modelDescription=description
+            modelDescription=description,
        )
-
--- a/py/utils/utils.py
+++ b/py/utils/utils.py
@@ -7,33 +7,47 @@ from ..config import config
 from ..services.settings_manager import get_settings_manager
 import asyncio

+
 def get_lora_info(lora_name):
    """Get the lora path and trigger words from cache"""
+
    async def _get_lora_info_async():
        scanner = await ServiceRegistry.get_lora_scanner()
        cache = await scanner.get_cached_data()
-        
+
        for item in cache.raw_data:
-            if item.get('file_name') == lora_name:
-                file_path = item.get('file_path')
+            if item.get("file_name") == lora_name:
+                file_path = item.get("file_path")
                if file_path:
-                    for root in config.loras_roots:
-                        root = root.replace(os.sep, '/')
+                    # Check all lora roots including extra paths
+                    all_roots = list(config.loras_roots or []) + list(
+                        config.extra_loras_roots or []
+                    )
+                    for root in all_roots:
+                        root = root.replace(os.sep, "/")
                        if file_path.startswith(root):
-                            relative_path = os.path.relpath(file_path, root).replace(os.sep, '/')
+                            relative_path = os.path.relpath(file_path, root).replace(
+                                os.sep, "/"
+                            )
                            # Get trigger words from civitai metadata
-                            civitai = item.get('civitai', {})
-                            trigger_words = civitai.get('trainedWords', []) if civitai else []
+                            civitai = item.get("civitai", {})
+                            trigger_words = (
+                                civitai.get("trainedWords", []) if civitai else []
+                            )
                            return relative_path, trigger_words
+                    # If not found in any root, return path with trigger words from cache
+                    civitai = item.get("civitai", {})
+                    trigger_words = civitai.get("trainedWords", []) if civitai else []
+                    return file_path, trigger_words
        return lora_name, []
-    
+
    try:
        # Check if we're already in an event loop
        loop = asyncio.get_running_loop()
        # If we're in a running loop, we need to use a different approach
        # Create a new thread to run the async code
        import concurrent.futures
-        
+
        def run_in_thread():
            new_loop = asyncio.new_event_loop()
            asyncio.set_event_loop(new_loop)
@@ -41,11 +55,11 @@ def get_lora_info(lora_name):
                return new_loop.run_until_complete(_get_lora_info_async())
            finally:
                new_loop.close()
-        
+
        with concurrent.futures.ThreadPoolExecutor() as executor:
            future = executor.submit(run_in_thread)
            return future.result()
-            
+
    except RuntimeError:
        # No event loop is running, we can use asyncio.run()
        return asyncio.run(_get_lora_info_async())
@@ -53,33 +67,34 @@ def get_lora_info(lora_name):

 def get_lora_info_absolute(lora_name):
    """Get the absolute lora path and trigger words from cache
-    
+
    Returns:
-        tuple: (absolute_path, trigger_words) where absolute_path is the full 
+        tuple: (absolute_path, trigger_words) where absolute_path is the full
               file system path to the LoRA file, or original lora_name if not found
    """
+
    async def _get_lora_info_absolute_async():
        scanner = await ServiceRegistry.get_lora_scanner()
        cache = await scanner.get_cached_data()
-        
+
        for item in cache.raw_data:
-            if item.get('file_name') == lora_name:
-                file_path = item.get('file_path')
+            if item.get("file_name") == lora_name:
+                file_path = item.get("file_path")
                if file_path:
                    # Return absolute path directly
                    # Get trigger words from civitai metadata
-                    civitai = item.get('civitai', {})
-                    trigger_words = civitai.get('trainedWords', []) if civitai else []
+                    civitai = item.get("civitai", {})
+                    trigger_words = civitai.get("trainedWords", []) if civitai else []
                    return file_path, trigger_words
        return lora_name, []
-    
+
    try:
        # Check if we're already in an event loop
        loop = asyncio.get_running_loop()
        # If we're in a running loop, we need to use a different approach
        # Create a new thread to run the async code
        import concurrent.futures
-        
+
        def run_in_thread():
            new_loop = asyncio.new_event_loop()
            asyncio.set_event_loop(new_loop)
@@ -87,50 +102,158 @@ def get_lora_info_absolute(lora_name):
                return new_loop.run_until_complete(_get_lora_info_absolute_async())
            finally:
                new_loop.close()
-        
+
        with concurrent.futures.ThreadPoolExecutor() as executor:
            future = executor.submit(run_in_thread)
            return future.result()
-            
+
    except RuntimeError:
        # No event loop is running, we can use asyncio.run()
        return asyncio.run(_get_lora_info_absolute_async())

-def fuzzy_match(text: str, pattern: str, threshold: float = 0.85) -> bool:
-        """
-        Check if text matches pattern using fuzzy matching.
-        Returns True if similarity ratio is above threshold.
-        """
-        if not pattern or not text:
-            return False
-        
-        # Convert both to lowercase for case-insensitive matching
-        text = text.lower()
-        pattern = pattern.lower()
-        
-        # Split pattern into words
-        search_words = pattern.split()
-        
-        # Check each word
-        for word in search_words:
-            # First check if word is a substring (faster)
-            if word in text:
+
+def get_checkpoint_info_absolute(checkpoint_name):
+    """Get the absolute checkpoint path and metadata from cache
+
+    Supports ComfyUI-style model names (e.g., "folder/model_name.ext")
+
+    Args:
+        checkpoint_name: The model name, can be:
+            - ComfyUI format: "folder/model_name.safetensors"
+            - Simple name: "model_name"
+
+    Returns:
+        tuple: (absolute_path, metadata) where absolute_path is the full
+               file system path to the checkpoint file, or original checkpoint_name if not found,
+               metadata is the full model metadata dict or None
+    """
+
+    async def _get_checkpoint_info_absolute_async():
+        from ..services.service_registry import ServiceRegistry
+
+        scanner = await ServiceRegistry.get_checkpoint_scanner()
+        cache = await scanner.get_cached_data()
+
+        # Get model roots for matching
+        model_roots = scanner.get_model_roots()
+
+        # Normalize the checkpoint name
+        normalized_name = checkpoint_name.replace(os.sep, "/")
+
+        for item in cache.raw_data:
+            file_path = item.get("file_path", "")
+            if not file_path:
                continue
-            
-            # If not found as substring, try fuzzy matching
-            # Check if any part of the text matches this word
-            found_match = False
-            for text_part in text.split():
-                ratio = SequenceMatcher(None, text_part, word).ratio()
-                if ratio >= threshold:
-                    found_match = True
-                    break
-                
-            if not found_match:
-                return False
-        
-        # All words found either as substrings or fuzzy matches
-        return True
+
+            # Format the stored path as ComfyUI-style name
+            formatted_name = _format_model_name_for_comfyui(file_path, model_roots)
+
+            # Match by formatted name
+            if formatted_name == normalized_name or formatted_name == checkpoint_name:
+                return file_path, item
+
+            # Also try matching by basename only (for backward compatibility)
+            file_name = item.get("file_name", "")
+            if (
+                file_name == checkpoint_name
+                or file_name == os.path.splitext(normalized_name)[0]
+            ):
+                return file_path, item
+
+        return checkpoint_name, None
+
+    try:
+        # Check if we're already in an event loop
+        loop = asyncio.get_running_loop()
+        # If we're in a running loop, we need to use a different approach
+        # Create a new thread to run the async code
+        import concurrent.futures
+
+        def run_in_thread():
+            new_loop = asyncio.new_event_loop()
+            asyncio.set_event_loop(new_loop)
+            try:
+                return new_loop.run_until_complete(
+                    _get_checkpoint_info_absolute_async()
+                )
+            finally:
+                new_loop.close()
+
+        with concurrent.futures.ThreadPoolExecutor() as executor:
+            future = executor.submit(run_in_thread)
+            return future.result()
+
+    except RuntimeError:
+        # No event loop is running, we can use asyncio.run()
+        return asyncio.run(_get_checkpoint_info_absolute_async())
+
+
+def _format_model_name_for_comfyui(file_path: str, model_roots: list) -> str:
+    """Format file path to ComfyUI-style model name (relative path with extension)
+
+    Example: /path/to/checkpoints/Illustrious/model.safetensors -> Illustrious/model.safetensors
+
+    Args:
+        file_path: Absolute path to the model file
+        model_roots: List of model root directories
+
+    Returns:
+        ComfyUI-style model name with relative path and extension
+    """
+    # Normalize path separators
+    normalized_path = file_path.replace(os.sep, "/")
+
+    # Find the matching root and get relative path
+    for root in model_roots:
+        normalized_root = root.replace(os.sep, "/")
+        # Ensure root ends with / for proper matching
+        if not normalized_root.endswith("/"):
+            normalized_root += "/"
+
+        if normalized_path.startswith(normalized_root):
+            rel_path = normalized_path[len(normalized_root) :]
+            return rel_path
+
+    # If no root matches, just return the basename with extension
+    return os.path.basename(file_path)
+
+
+def fuzzy_match(text: str, pattern: str, threshold: float = 0.85) -> bool:
+    """
+    Check if text matches pattern using fuzzy matching.
+    Returns True if similarity ratio is above threshold.
+    """
+    if not pattern or not text:
+        return False
+
+    # Convert both to lowercase for case-insensitive matching
+    text = text.lower()
+    pattern = pattern.lower()
+
+    # Split pattern into words
+    search_words = pattern.split()
+
+    # Check each word
+    for word in search_words:
+        # First check if word is a substring (faster)
+        if word in text:
+            continue
+
+        # If not found as substring, try fuzzy matching
+        # Check if any part of the text matches this word
+        found_match = False
+        for text_part in text.split():
+            ratio = SequenceMatcher(None, text_part, word).ratio()
+            if ratio >= threshold:
+                found_match = True
+                break
+
+        if not found_match:
+            return False
+
+    # All words found either as substrings or fuzzy matches
+    return True
+

 def sanitize_folder_name(name: str, replacement: str = "_") -> str:
    """Sanitize a folder name by removing or replacing invalid characters.
@@ -156,10 +279,13 @@ def sanitize_folder_name(name: str, replacement: str = "_") -> str:
    # Collapse repeated replacement characters to a single instance
    if replacement:
        sanitized = re.sub(f"{re.escape(replacement)}+", replacement, sanitized)
-        sanitized = sanitized.strip(replacement)
-
-    # Remove trailing spaces or periods which are invalid on Windows
-    sanitized = sanitized.rstrip(" .")
+        # Combine stripping to be idempotent:
+        # Right side: strip replacement, space, and dot (Windows restriction)
+        # Left side: strip replacement and space (leading dots are allowed)
+        sanitized = sanitized.rstrip(" ." + replacement).lstrip(" " + replacement)
+    else:
+        # If no replacement, just strip spaces and dots from right, spaces from left
+        sanitized = sanitized.rstrip(" .").lstrip(" ")

    if not sanitized:
        return "unnamed"
@@ -170,25 +296,25 @@ def sanitize_folder_name(name: str, replacement: str = "_") -> str:
 def calculate_recipe_fingerprint(loras):
    """
    Calculate a unique fingerprint for a recipe based on its LoRAs.
-    
+
    The fingerprint is created by sorting LoRA hashes, filtering invalid entries,
    normalizing strength values to 2 decimal places, and joining in format:
    hash1:strength1|hash2:strength2|...
-    
+
    Args:
        loras (list): List of LoRA dictionaries with hash and strength values
-        
+
    Returns:
        str: The calculated fingerprint
    """
    if not loras:
        return ""
-    
+
    valid_loras = []
    for lora in loras:
        if lora.get("exclude", False):
            continue
-            
+
        hash_value = lora.get("hash", "")
        if isinstance(hash_value, str):
            hash_value = hash_value.lower()
@@ -206,18 +332,23 @@ def calculate_recipe_fingerprint(loras):
            strength = round(float(strength_val), 2)
        except (ValueError, TypeError):
            strength = 1.0
-        
+
        valid_loras.append((hash_value, strength))
-    
+
    # Sort by hash
    valid_loras.sort()
-    
+
    # Join in format hash1:strength1|hash2:strength2|...
-    fingerprint = "|".join([f"{hash_value}:{strength}" for hash_value, strength in valid_loras])
-    
+    fingerprint = "|".join(
+        [f"{hash_value}:{strength}" for hash_value, strength in valid_loras]
+    )
+
    return fingerprint

-def calculate_relative_path_for_model(model_data: Dict, model_type: str = 'lora') -> str:
+
+def calculate_relative_path_for_model(
+    model_data: Dict, model_type: str = "lora"
+) -> str:
    """Calculate relative path for existing model using template from settings

    Args:
@@ -233,77 +364,80 @@ def calculate_relative_path_for_model(model_data: Dict, model_type: str = 'lora'

    # If template is empty, return empty path (flat structure)
    if not path_template:
-        return ''
+        return ""

    # Get base model name from model metadata
-    civitai_data = model_data.get('civitai', {})
+    civitai_data = model_data.get("civitai", {})

    # For CivitAI models, prefer civitai data only if 'id' exists; for non-CivitAI models, use model_data directly
-    if civitai_data and civitai_data.get('id') is not None:
-        base_model = model_data.get('base_model', '')
+    if civitai_data and civitai_data.get("id") is not None:
+        base_model = model_data.get("base_model", "")
        # Get author from civitai creator data
-        creator_info = civitai_data.get('creator') or {}
-        author = creator_info.get('username') or 'Anonymous'
+        creator_info = civitai_data.get("creator") or {}
+        author = creator_info.get("username") or "Anonymous"
    else:
        # Fallback to model_data fields for non-CivitAI models
-        base_model = model_data.get('base_model', '')
-        author = 'Anonymous'  # Default for non-CivitAI models
+        base_model = model_data.get("base_model", "")
+        author = "Anonymous"  # Default for non-CivitAI models

-    model_tags = model_data.get('tags', [])
+    model_tags = model_data.get("tags", [])

    # Apply mapping if available
-    base_model_mappings = settings_manager.get('base_model_path_mappings', {})
+    base_model_mappings = settings_manager.get("base_model_path_mappings", {})
    mapped_base_model = base_model_mappings.get(base_model, base_model)

    # Convert all tags to lowercase to avoid case sensitivity issues on Windows
    lowercase_tags = [tag.lower() for tag in model_tags if isinstance(tag, str)]
-    first_tag = settings_manager.resolve_priority_tag_for_model(lowercase_tags, model_type)
+    first_tag = settings_manager.resolve_priority_tag_for_model(
+        lowercase_tags, model_type
+    )

    if not first_tag:
-        first_tag = 'no tags'  # Default if no tags available
+        first_tag = "no tags"  # Default if no tags available

    # Format the template with available data
-    model_name = sanitize_folder_name(model_data.get('model_name', ''))
-    version_name = ''
+    model_name = sanitize_folder_name(model_data.get("model_name", ""))
+    version_name = ""

    if isinstance(civitai_data, dict):
-        version_name = sanitize_folder_name(civitai_data.get('name') or '')
+        version_name = sanitize_folder_name(civitai_data.get("name") or "")

    formatted_path = path_template
-    formatted_path = formatted_path.replace('{base_model}', mapped_base_model)
-    formatted_path = formatted_path.replace('{first_tag}', first_tag)
-    formatted_path = formatted_path.replace('{author}', author)
-    formatted_path = formatted_path.replace('{model_name}', model_name)
-    formatted_path = formatted_path.replace('{version_name}', version_name)
+    formatted_path = formatted_path.replace("{base_model}", mapped_base_model)
+    formatted_path = formatted_path.replace("{first_tag}", first_tag)
+    formatted_path = formatted_path.replace("{author}", author)
+    formatted_path = formatted_path.replace("{model_name}", model_name)
+    formatted_path = formatted_path.replace("{version_name}", version_name)

-    if model_type == 'embedding':
-        formatted_path = formatted_path.replace(' ', '_')
+    if model_type == "embedding":
+        formatted_path = formatted_path.replace(" ", "_")

    return formatted_path

+
 def remove_empty_dirs(path):
    """Recursively remove empty directories starting from the given path.
-    
+
    Args:
        path (str): Root directory to start cleaning from
-        
+
    Returns:
        int: Number of empty directories removed
    """
    removed_count = 0
-    
+
    if not os.path.isdir(path):
        return removed_count
-        
+
    # List all files in directory
    files = os.listdir(path)
-    
+
    # Process all subdirectories first
    for file in files:
        full_path = os.path.join(path, file)
        if os.path.isdir(full_path):
            removed_count += remove_empty_dirs(full_path)
-    
+
    # Check if directory is now empty (after processing subdirectories)
    if not os.listdir(path):
        try:
@@ -311,5 +445,5 @@ def remove_empty_dirs(path):
            removed_count += 1
        except OSError:
            pass
-    
+
    return removed_count
--- a/standalone.py
+++ b/standalone.py
@@ -345,6 +345,7 @@ class StandaloneLoraManager(LoraManager):
            "/ws/download-progress", ws_manager.handle_download_connection
        )
        app.router.add_get("/ws/init-progress", ws_manager.handle_init_connection)
+        app.router.add_get("/ws/batch-import-progress", ws_manager.handle_connection)

        # Schedule service initialization
        app.on_startup.append(lambda app: cls._initialize_services())
--- a/static/css/components/batch-import-modal.css
+++ b/static/css/components/batch-import-modal.css
@@ -0,0 +1,677 @@
+/* Batch Import Modal Styles */
+
+/* Step Containers */
+.batch-import-step {
+    margin: var(--space-2) 0;
+}
+
+/* Section Description */
+.section-description {
+    color: var(--text-color);
+    opacity: 0.8;
+    margin-bottom: var(--space-2);
+    font-size: 0.95em;
+}
+
+/* Hint Text */
+.input-hint {
+    display: flex;
+    align-items: center;
+    gap: 6px;
+    color: var(--text-color);
+    opacity: 0.7;
+    font-size: 0.85em;
+    margin-top: 6px;
+}
+
+.input-hint i {
+    color: var(--lora-accent);
+}
+
+/* Textarea Styling */
+#batchUrlInput {
+    width: 100%;
+    min-height: 120px;
+    padding: 12px;
+    border: 1px solid var(--border-color);
+    border-radius: var(--border-radius-xs);
+    background: var(--bg-color);
+    color: var(--text-color);
+    font-family: inherit;
+    font-size: 0.9em;
+    resize: vertical;
+    transition: border-color 0.2s, box-shadow 0.2s;
+}
+
+#batchUrlInput:focus {
+    outline: none;
+    border-color: var(--lora-accent);
+    box-shadow: 0 0 0 2px oklch(from var(--lora-accent) l c h / 0.2);
+}
+
+/* Checkbox Group */
+.checkbox-group {
+    margin-top: var(--space-2);
+}
+
+.checkbox-label {
+    display: flex;
+    align-items: center;
+    gap: 10px;
+    cursor: pointer;
+    color: var(--text-color);
+    font-size: 0.95em;
+    user-select: none;
+}
+
+.checkbox-label input[type="checkbox"] {
+    display: none;
+}
+
+.checkmark {
+    width: 18px;
+    height: 18px;
+    border: 2px solid var(--border-color);
+    border-radius: 4px;
+    display: flex;
+    align-items: center;
+    justify-content: center;
+    transition: all 0.2s;
+    background: var(--bg-color);
+}
+
+.checkbox-label input[type="checkbox"]:checked + .checkmark {
+    background: var(--lora-accent);
+    border-color: var(--lora-accent);
+}
+
+.checkbox-label input[type="checkbox"]:checked + .checkmark::after {
+    content: '\f00c';
+    font-family: 'Font Awesome 6 Free';
+    font-weight: 900;
+    color: var(--lora-text);
+    font-size: 12px;
+}
+
+/* Batch Options */
+.batch-options {
+    margin-top: var(--space-3);
+    padding-top: var(--space-3);
+    border-top: 1px solid var(--border-color);
+}
+
+/* Input with Button */
+.input-with-button {
+    display: flex;
+    gap: 8px;
+}
+
+.input-with-button input {
+    flex: 1;
+    min-width: 0;
+}
+
+.input-with-button button {
+    flex-shrink: 0;
+    white-space: nowrap;
+    padding: 8px 16px;
+    background: var(--lora-accent);
+    color: var(--lora-text);
+    border: none;
+    border-radius: var(--border-radius-xs);
+    cursor: pointer;
+    transition: background-color 0.2s;
+}
+
+.input-with-button button:hover {
+    background: oklch(from var(--lora-accent) l c h / 0.9);
+}
+
+/* Dark theme adjustments for input-with-button */
+[data-theme="dark"] .input-with-button button {
+    background: var(--lora-accent);
+    color: var(--lora-text);
+}
+
+[data-theme="dark"] .input-with-button button:hover {
+    background: oklch(from var(--lora-accent) calc(l - 0.1) c h);
+}
+
+/* Directory Browser */
+.directory-browser {
+    margin-top: var(--space-3);
+    border: 1px solid var(--border-color);
+    border-radius: var(--border-radius-xs);
+    background: var(--lora-surface);
+    overflow: hidden;
+}
+
+.browser-header {
+    display: flex;
+    align-items: center;
+    gap: 10px;
+    padding: 10px 12px;
+    background: var(--bg-color);
+    border-bottom: 1px solid var(--border-color);
+}
+
+.back-btn {
+    display: flex;
+    align-items: center;
+    justify-content: center;
+    width: 32px;
+    height: 32px;
+    border: 1px solid var(--border-color);
+    border-radius: var(--border-radius-xs);
+    background: var(--card-bg);
+    color: var(--text-color);
+    cursor: pointer;
+    transition: all 0.2s;
+}
+
+.back-btn:hover {
+    border-color: var(--lora-accent);
+    background: var(--bg-color);
+}
+
+.back-btn:disabled {
+    opacity: 0.5;
+    cursor: not-allowed;
+}
+
+.current-path {
+    flex: 1;
+    padding: 6px 10px;
+    background: var(--card-bg);
+    border: 1px solid var(--border-color);
+    border-radius: var(--border-radius-xs);
+    font-size: 0.9em;
+    color: var(--text-color);
+    white-space: nowrap;
+    overflow: hidden;
+    text-overflow: ellipsis;
+}
+
+.browser-content {
+    max-height: 300px;
+    overflow-y: auto;
+    padding: 12px;
+}
+
+.browser-section {
+    margin-bottom: 16px;
+}
+
+.browser-section:last-child {
+    margin-bottom: 0;
+}
+
+.section-label {
+    display: flex;
+    align-items: center;
+    gap: 8px;
+    font-weight: 600;
+    font-size: 0.85em;
+    color: var(--text-color);
+    margin-bottom: 8px;
+    padding-bottom: 6px;
+    border-bottom: 1px solid var(--border-color);
+}
+
+.section-label i {
+    color: var(--lora-accent);
+}
+
+.folder-list,
+.file-list {
+    display: flex;
+    flex-direction: column;
+    gap: 4px;
+}
+
+.folder-item,
+.file-item {
+    display: flex;
+    align-items: center;
+    gap: 10px;
+    padding: 8px 10px;
+    border-radius: var(--border-radius-xs);
+    cursor: pointer;
+    transition: all 0.2s;
+    border: 1px solid transparent;
+}
+
+.folder-item:hover,
+.file-item:hover {
+    background: var(--lora-surface-hover, oklch(from var(--lora-accent) l c h / 0.1));
+    border-color: var(--lora-accent);
+}
+
+.folder-item.selected,
+.file-item.selected {
+    background: oklch(from var(--lora-accent) l c h / 0.15);
+    border-color: var(--lora-accent);
+}
+
+.folder-item i {
+    color: #fbbf24;
+    font-size: 1.1em;
+}
+
+.file-item i {
+    color: var(--text-color);
+    opacity: 0.6;
+    font-size: 1em;
+}
+
+.item-name {
+    flex: 1;
+    font-size: 0.9em;
+    color: var(--text-color);
+    white-space: nowrap;
+    overflow: hidden;
+    text-overflow: ellipsis;
+}
+
+.item-size {
+    font-size: 0.8em;
+    color: var(--text-color);
+    opacity: 0.6;
+}
+
+.browser-footer {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    padding: 10px 12px;
+    background: var(--bg-color);
+    border-top: 1px solid var(--border-color);
+}
+
+.stats {
+    font-size: 0.85em;
+    color: var(--text-color);
+    opacity: 0.8;
+}
+
+.stats span {
+    font-weight: 600;
+    color: var(--lora-accent);
+}
+
+/* Dark theme adjustments */
+[data-theme="dark"] .directory-browser {
+    background: var(--card-bg);
+}
+
+[data-theme="dark"] .browser-header,
+[data-theme="dark"] .browser-footer {
+    background: var(--lora-surface);
+}
+
+[data-theme="dark"] .folder-item i {
+    color: #fcd34d;
+}
+
+/* Progress Container */
+.batch-progress-container {
+    padding: var(--space-3);
+    background: var(--lora-surface);
+    border-radius: var(--border-radius-sm);
+    margin-bottom: var(--space-3);
+}
+
+.progress-header {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    margin-bottom: var(--space-2);
+}
+
+.progress-status {
+    display: flex;
+    align-items: center;
+    gap: 10px;
+}
+
+.status-icon {
+    color: var(--lora-accent);
+    font-size: 1.1em;
+}
+
+.status-icon i {
+    animation: fa-spin 2s infinite linear;
+}
+
+.status-text {
+    font-weight: 500;
+    color: var(--text-color);
+}
+
+.progress-percentage {
+    font-size: 1.2em;
+    font-weight: 600;
+    color: var(--lora-accent);
+}
+
+/* Progress Bar */
+.progress-bar-container {
+    height: 8px;
+    background: var(--bg-color);
+    border-radius: 4px;
+    overflow: hidden;
+    margin-bottom: var(--space-3);
+}
+
+.progress-bar {
+    height: 100%;
+    background: linear-gradient(90deg, var(--lora-accent), oklch(from var(--lora-accent) calc(l + 0.1) c h));
+    border-radius: 4px;
+    transition: width 0.3s ease;
+}
+
+/* Progress Stats */
+.progress-stats {
+    display: grid;
+    grid-template-columns: repeat(4, 1fr);
+    gap: var(--space-2);
+    margin-bottom: var(--space-2);
+}
+
+.stat-item {
+    display: flex;
+    flex-direction: column;
+    align-items: center;
+    padding: var(--space-2);
+    background: var(--bg-color);
+    border-radius: var(--border-radius-xs);
+    border: 1px solid var(--border-color);
+}
+
+.stat-item.success {
+    border-left: 3px solid #00B87A;
+}
+
+.stat-item.failed {
+    border-left: 3px solid var(--lora-error);
+}
+
+.stat-item.skipped {
+    border-left: 3px solid var(--lora-warning);
+}
+
+.stat-label {
+    font-size: 0.8em;
+    color: var(--text-color);
+    opacity: 0.7;
+    margin-bottom: 4px;
+}
+
+.stat-value {
+    font-size: 1.4em;
+    font-weight: 600;
+    color: var(--text-color);
+}
+
+/* Current Item */
+.current-item {
+    display: flex;
+    align-items: baseline;
+    gap: 10px;
+    padding: var(--space-2);
+    background: var(--bg-color);
+    border-radius: var(--border-radius-xs);
+    font-size: 0.9em;
+}
+
+.current-item-label {
+    color: var(--text-color);
+    opacity: 0.7;
+    flex-shrink: 0;
+}
+
+.current-item-name {
+    color: var(--text-color);
+    font-weight: 500;
+    flex: 1;
+    white-space: nowrap;
+    overflow: hidden;
+    text-overflow: ellipsis;
+    line-height: 1.2;
+}
+
+/* Results Container */
+.batch-results-container {
+    padding: var(--space-3);
+    background: var(--lora-surface);
+    border-radius: var(--border-radius-sm);
+    margin-bottom: var(--space-3);
+}
+
+.results-header {
+    text-align: center;
+    margin-bottom: var(--space-3);
+}
+
+.results-icon {
+    font-size: 3em;
+    color: #00B87A;
+    margin-bottom: var(--space-1);
+}
+
+.results-icon.warning {
+    color: var(--lora-warning);
+}
+
+.results-icon.error {
+    color: var(--lora-error);
+}
+
+.results-title {
+    font-size: 1.3em;
+    font-weight: 600;
+    color: var(--text-color);
+}
+
+/* Results Summary - Matches progress-stats styling */
+.results-summary {
+    display: grid;
+    grid-template-columns: repeat(4, 1fr);
+    gap: var(--space-2);
+    margin-bottom: var(--space-3);
+}
+
+.result-card {
+    display: flex;
+    flex-direction: column;
+    align-items: center;
+    padding: var(--space-2);
+    background: var(--bg-color);
+    border-radius: var(--border-radius-xs);
+    border: 1px solid var(--border-color);
+    text-align: center;
+}
+
+.result-card.success {
+    border-left: 3px solid #00B87A;
+}
+
+.result-card.failed {
+    border-left: 3px solid var(--lora-error);
+}
+
+.result-card.skipped {
+    border-left: 3px solid var(--lora-warning);
+}
+
+.result-label {
+    font-size: 0.8em;
+    color: var(--text-color);
+    opacity: 0.7;
+    margin-bottom: 4px;
+}
+
+.result-value {
+    font-size: 1.4em;
+    font-weight: 600;
+    color: var(--text-color);
+}
+
+/* Results Details */
+.results-details {
+    border-top: 1px solid var(--border-color);
+    padding-top: var(--space-2);
+}
+
+.details-toggle {
+    display: flex;
+    align-items: center;
+    justify-content: center;
+    gap: 8px;
+    padding: 10px;
+    cursor: pointer;
+    color: var(--lora-accent);
+    font-weight: 500;
+    border-radius: var(--border-radius-xs);
+    transition: background 0.2s;
+}
+
+.details-toggle:hover {
+    background: oklch(from var(--lora-accent) l c h / 0.1);
+}
+
+.details-toggle i {
+    transition: transform 0.2s;
+}
+
+.details-toggle.expanded i {
+    transform: rotate(180deg);
+}
+
+.details-list {
+    max-height: 250px;
+    overflow-y: auto;
+    margin-top: var(--space-2);
+    background: var(--bg-color);
+    border-radius: var(--border-radius-xs);
+    border: 1px solid var(--border-color);
+}
+
+/* Result Item in Details */
+.result-item {
+    display: flex;
+    align-items: center;
+    gap: 10px;
+    padding: 10px 12px;
+    border-bottom: 1px solid var(--border-color);
+    font-size: 0.9em;
+}
+
+.result-item:last-child {
+    border-bottom: none;
+}
+
+.result-item-status {
+    width: 24px;
+    height: 24px;
+    border-radius: 50%;
+    display: flex;
+    align-items: center;
+    justify-content: center;
+    font-size: 0.8em;
+}
+
+.result-item-status.success {
+    background: oklch(from #00B87A l c h / 0.2);
+    color: #00B87A;
+}
+
+.result-item-status.failed {
+    background: oklch(from var(--lora-error) l c h / 0.2);
+    color: var(--lora-error);
+}
+
+.result-item-status.skipped {
+    background: oklch(from var(--lora-warning) l c h / 0.2);
+    color: var(--lora-warning);
+}
+
+.result-item-info {
+    flex: 1;
+    min-width: 0;
+}
+
+.result-item-name {
+    font-weight: 500;
+    color: var(--text-color);
+    white-space: nowrap;
+    overflow: hidden;
+    text-overflow: ellipsis;
+}
+
+.result-item-error {
+    font-size: 0.8em;
+    color: var(--lora-error);
+    margin-top: 2px;
+}
+
+/* Responsive Adjustments */
+@media (max-width: 768px) {
+    .progress-stats,
+    .results-summary {
+        grid-template-columns: repeat(2, 1fr);
+    }
+    
+    .batch-progress-container,
+    .batch-results-container {
+        padding: var(--space-2);
+    }
+}
+
+/* Dark Theme Adjustments */
+[data-theme="dark"] .batch-progress-container,
+[data-theme="dark"] .batch-results-container {
+    background: var(--card-bg);
+}
+
+[data-theme="dark"] .stat-item,
+[data-theme="dark"] .result-card,
+[data-theme="dark"] .current-item,
+[data-theme="dark"] .details-list {
+    background: var(--lora-surface);
+}
+
+/* Cancelled State */
+.batch-progress-container.cancelled .progress-bar {
+    background: var(--lora-warning);
+}
+
+.batch-progress-container.cancelled .status-icon {
+    color: var(--lora-warning);
+}
+
+/* Error State */
+.batch-progress-container.error .progress-bar {
+    background: var(--lora-error);
+}
+
+.batch-progress-container.error .status-icon {
+    color: var(--lora-error);
+}
+
+/* Completed State */
+.batch-progress-container.completed .progress-bar {
+    background: #00B87A;
+}
+
+.batch-progress-container.completed .status-icon {
+    color: #00B87A;
+}
+
+.batch-progress-container.completed .status-icon i {
+    animation: none;
+}
+
+.batch-progress-container.completed .status-icon i::before {
+    content: '\f00c';
+}
--- a/static/css/components/menu.css
+++ b/static/css/components/menu.css
@@ -130,7 +130,7 @@
    max-height: 400px;
    overflow-y: auto;
    box-shadow: 0 2px 10px rgba(0, 0, 0, 0.2);
-    z-index: 1000;
+    z-index: var(--z-overlay);
    display: none;
    backdrop-filter: blur(10px);
 }
--- a/static/js/api/recipeApi.js
+++ b/static/js/api/recipeApi.js
@@ -259,6 +259,26 @@ export async function resetAndReload(updateFolders = false) {
    });
 }

+/**
+ * Sync changes - quick refresh without rebuilding cache (similar to models page)
+ */
+export async function syncChanges() {
+    try {
+        state.loadingManager.showSimpleLoading('Syncing changes...');
+
+        // Simply reload the recipes without rebuilding cache
+        await resetAndReload();
+
+        showToast('toast.recipes.syncComplete', {}, 'success');
+    } catch (error) {
+        console.error('Error syncing recipes:', error);
+        showToast('toast.recipes.syncFailed', { message: error.message }, 'error');
+    } finally {
+        state.loadingManager.hide();
+        state.loadingManager.restoreProgressBar();
+    }
+}
+
 /**
 * Refreshes the recipe list by first rebuilding the cache and then loading recipes
 */
--- a/static/js/components/ContextMenu/BulkContextMenu.js
+++ b/static/js/components/ContextMenu/BulkContextMenu.js
@@ -117,7 +117,10 @@ export class BulkContextMenu extends BaseContextMenu {
    countSkipStatus(skipState) {
        let count = 0;
        for (const filePath of state.selectedModels) {
-            const card = document.querySelector(`.model-card[data-filepath="${filePath}"]`);
+            const escapedPath = window.CSS && typeof window.CSS.escape === 'function'
+                ? window.CSS.escape(filePath)
+                : filePath.replace(/["\\]/g, '\\$&');
+            const card = document.querySelector(`.model-card[data-filepath="${escapedPath}"]`);
            if (card) {
                const isSkipped = card.dataset.skip_metadata_refresh === 'true';
                if (isSkipped === skipState) {
--- a/static/js/components/RecipeCard.js
+++ b/static/js/components/RecipeCard.js
@@ -201,8 +201,9 @@ class RecipeCard {
            this.recipe.favorite = isFavorite;

            // Re-find star icon in case of re-render during fault
+            const filePathForXpath = this.recipe.file_path.replace(/"/g, '&quot;');
            const currentCard = card.ownerDocument.evaluate(
-                `.//*[@data-filepath="${this.recipe.file_path}"]`,
+                `.//*[@data-filepath="${filePathForXpath}"]`,
                card.ownerDocument, null, XPathResult.FIRST_ORDERED_NODE_TYPE, null
            ).singleNodeValue || card;

--- a/static/js/components/SidebarManager.js
+++ b/static/js/components/SidebarManager.js
@@ -7,6 +7,7 @@ import { translate } from '../utils/i18nHelpers.js';
 import { state } from '../state/index.js';
 import { bulkManager } from '../managers/BulkManager.js';
 import { showToast } from '../utils/uiHelpers.js';
+import { escapeHtml, escapeAttribute } from './shared/utils.js';

 export class SidebarManager {
    constructor() {
@@ -1294,15 +1295,19 @@ export class SidebarManager {
            const isExpanded = this.expandedNodes.has(currentPath);
            const isSelected = this.selectedPath === currentPath;

+            const escapedPath = escapeAttribute(currentPath);
+            const escapedFolderName = escapeHtml(folderName);
+            const escapedTitle = escapeAttribute(folderName);
+
            return `
-                <div class="sidebar-tree-node" data-path="${currentPath}">
-                    <div class="sidebar-tree-node-content ${isSelected ? 'selected' : ''}" data-path="${currentPath}">
+                <div class="sidebar-tree-node" data-path="${escapedPath}">
+                    <div class="sidebar-tree-node-content ${isSelected ? 'selected' : ''}" data-path="${escapedPath}">
                        <div class="sidebar-tree-expand-icon ${isExpanded ? 'expanded' : ''}" 
                             style="${hasChildren ? '' : 'opacity: 0; pointer-events: none;'}">
                            <i class="fas fa-chevron-right"></i>
                        </div>
                        <i class="fas fa-folder sidebar-tree-folder-icon"></i>
-                        <div class="sidebar-tree-folder-name" title="${folderName}">${folderName}</div>
+                        <div class="sidebar-tree-folder-name" title="${escapedTitle}">${escapedFolderName}</div>
                    </div>
                    ${hasChildren ? `
                        <div class="sidebar-tree-children ${isExpanded ? 'expanded' : ''}">
@@ -1342,12 +1347,15 @@ export class SidebarManager {
        const foldersHtml = this.foldersList.map(folder => {
            const displayName = folder === '' ? '/' : folder;
            const isSelected = this.selectedPath === folder;
+            const escapedPath = escapeAttribute(folder);
+            const escapedDisplayName = escapeHtml(displayName);
+            const escapedTitle = escapeAttribute(displayName);

            return `
-                <div class="sidebar-folder-item ${isSelected ? 'selected' : ''}" data-path="${folder}">
-                    <div class="sidebar-node-content" data-path="${folder}">
+                <div class="sidebar-folder-item ${isSelected ? 'selected' : ''}" data-path="${escapedPath}">
+                    <div class="sidebar-node-content" data-path="${escapedPath}">
                        <i class="fas fa-folder sidebar-folder-icon"></i>
-                        <div class="sidebar-folder-name" title="${displayName}">${displayName}</div>
+                        <div class="sidebar-folder-name" title="${escapedTitle}">${escapedDisplayName}</div>
                    </div>
                </div>
            `;
@@ -1570,7 +1578,8 @@ export class SidebarManager {

            // Add selection to current path
            if (this.selectedPath !== null && this.selectedPath !== undefined) {
-                const selectedItem = folderTree.querySelector(`[data-path="${this.selectedPath}"]`);
+                const escapedPathSelector = CSS.escape(this.selectedPath);
+                const selectedItem = folderTree.querySelector(`[data-path="${escapedPathSelector}"]`);
                if (selectedItem) {
                    selectedItem.classList.add('selected');
                }
@@ -1581,7 +1590,8 @@ export class SidebarManager {
            });

            if (this.selectedPath !== null && this.selectedPath !== undefined) {
-                const selectedNode = folderTree.querySelector(`[data-path="${this.selectedPath}"] .sidebar-tree-node-content`);
+                const escapedPathSelector = CSS.escape(this.selectedPath);
+                const selectedNode = folderTree.querySelector(`[data-path="${escapedPathSelector}"] .sidebar-tree-node-content`);
                if (selectedNode) {
                    selectedNode.classList.add('selected');
                    this.expandPathParents(this.selectedPath);
@@ -1655,7 +1665,7 @@ export class SidebarManager {
        const breadcrumbs = [`
            <div class="breadcrumb-dropdown">
                <span class="sidebar-breadcrumb-item ${isRootSelected ? 'active' : ''}" data-path="">
-                    <i class="fas fa-home"></i> ${this.apiClient.apiConfig.config.displayName} root
+                    <i class="fas fa-home"></i> ${escapeHtml(this.apiClient.apiConfig.config.displayName)} root
                </span>
            </div>
        `];
@@ -1675,8 +1685,8 @@ export class SidebarManager {
                        </span>
                        <div class="breadcrumb-dropdown-menu">
                            ${nextLevelFolders.map(folder => `
-                                <div class="breadcrumb-dropdown-item" data-path="${folder}">
-                                    ${folder}
+                                <div class="breadcrumb-dropdown-item" data-path="${escapeAttribute(folder)}">
+                                    ${escapeHtml(folder)}
                                </div>`).join('')
                    }
                        </div>
@@ -1692,12 +1702,14 @@ export class SidebarManager {

            // Get siblings for this level
            const siblings = this.getSiblingFolders(parts, index);
+            const escapedCurrentPath = escapeAttribute(currentPath);
+            const escapedPart = escapeHtml(part);

            breadcrumbs.push(`<span class="sidebar-breadcrumb-separator">/</span>`);
            breadcrumbs.push(`
                <div class="breadcrumb-dropdown">
-                    <span class="sidebar-breadcrumb-item ${isLast ? 'active' : ''}" data-path="${currentPath}">
-                        ${part}
+                    <span class="sidebar-breadcrumb-item ${isLast ? 'active' : ''}" data-path="${escapedCurrentPath}">
+                        ${escapedPart}
                        ${siblings.length > 1 ? `
                            <span class="breadcrumb-dropdown-indicator">
                                <i class="fas fa-caret-down"></i>
@@ -1706,11 +1718,14 @@ export class SidebarManager {
                    </span>
                    ${siblings.length > 1 ? `
                        <div class="breadcrumb-dropdown-menu">
-                            ${siblings.map(folder => `
-                                <div class="breadcrumb-dropdown-item ${folder === part ? 'active' : ''}" 
-                                     data-path="${currentPath.replace(part, folder)}">
-                                    ${folder}
-                                </div>`).join('')
+                            ${siblings.map(folder => {
+                                const siblingPath = parts.slice(0, index).concat(folder).join('/');
+                                return `
+                                    <div class="breadcrumb-dropdown-item ${folder === part ? 'active' : ''}" 
+                                         data-path="${escapeAttribute(siblingPath)}">
+                                        ${escapeHtml(folder)}
+                                    </div>`;
+                            }).join('')
                    }
                        </div>
                    ` : ''}
@@ -1732,8 +1747,8 @@ export class SidebarManager {
                            </span>
                            <div class="breadcrumb-dropdown-menu">
                                ${childFolders.map(folder => `
-                                    <div class="breadcrumb-dropdown-item" data-path="${currentPath}/${folder}">
-                                        ${folder}
+                                    <div class="breadcrumb-dropdown-item" data-path="${escapeAttribute(currentPath + '/' + folder)}">
+                                        ${escapeHtml(folder)}
                                    </div>`).join('')
                        }
                            </div>
--- a/static/js/components/shared/ModelModal.js
+++ b/static/js/components/shared/ModelModal.js
@@ -846,8 +846,14 @@ function setupLoraSpecificFields(filePath) {

        const currentPath = resolveFilePath();
        if (!currentPath) return;
-        const loraCard = document.querySelector(`.model-card[data-filepath="${currentPath}"]`) ||
-            document.querySelector(`.model-card[data-filepath="${filePath}"]`);
+        const escapedCurrentPath = window.CSS && typeof window.CSS.escape === 'function'
+            ? window.CSS.escape(currentPath)
+            : currentPath.replace(/["\\]/g, '\\$&');
+        const escapedFilePath = window.CSS && typeof window.CSS.escape === 'function'
+            ? window.CSS.escape(filePath)
+            : filePath.replace(/["\\]/g, '\\$&');
+        const loraCard = document.querySelector(`.model-card[data-filepath="${escapedCurrentPath}"]`) ||
+            document.querySelector(`.model-card[data-filepath="${escapedFilePath}"]`);
        const currentPresets = parsePresets(loraCard?.dataset.usage_tips);

        if (key === 'strength_range') {
--- a/static/js/components/shared/PresetTags.js
+++ b/static/js/components/shared/PresetTags.js
@@ -49,7 +49,10 @@ function formatPresetKey(key) {
 */
 window.removePreset = async function(key) {
    const filePath = document.querySelector('#modelModal .modal-content .file-path').dataset.filepath;
-    const loraCard = document.querySelector(`.model-card[data-filepath="${filePath}"]`);
+    const escapedPath = window.CSS && typeof window.CSS.escape === 'function'
+        ? window.CSS.escape(filePath)
+        : filePath.replace(/["\\]/g, '\\$&');
+    const loraCard = document.querySelector(`.model-card[data-filepath="${escapedPath}"]`);
    const currentPresets = parsePresets(loraCard.dataset.usage_tips);
    
    delete currentPresets[key];
--- a/static/js/components/shared/showcase/MetadataPanel.js
+++ b/static/js/components/shared/showcase/MetadataPanel.js
@@ -2,6 +2,7 @@
 * MetadataPanel.js
 * Generates metadata panels for showcase media items
 */
+import { escapeHtml } from '../utils.js';

 /**
 * Generate metadata panel HTML
@@ -49,6 +50,7 @@ export function generateMetadataPanel(hasParams, hasPrompts, prompt, negativePro
    }
    
    if (prompt) {
+        prompt = escapeHtml(prompt);
        content += `
            <div class="metadata-row prompt-row">
                <span class="metadata-label">Prompt:</span>
@@ -64,6 +66,7 @@ export function generateMetadataPanel(hasParams, hasPrompts, prompt, negativePro
    }
    
    if (negativePrompt) {
+        negativePrompt = escapeHtml(negativePrompt);
        content += `
            <div class="metadata-row prompt-row">
                <span class="metadata-label">Negative Prompt:</span>
@@ -80,4 +83,4 @@ export function generateMetadataPanel(hasParams, hasPrompts, prompt, negativePro
    
    content += '</div></div>';
    return content;
-}
+}
--- a/static/js/managers/BatchImportManager.js
+++ b/static/js/managers/BatchImportManager.js
@@ -0,0 +1,807 @@
+import { modalManager } from './ModalManager.js';
+import { showToast } from '../utils/uiHelpers.js';
+import { translate } from '../utils/i18nHelpers.js';
+import { WS_ENDPOINTS } from '../api/apiConfig.js';
+import { getStorageItem, setStorageItem } from '../utils/storageHelpers.js';
+
+/**
+ * Manager for batch importing recipes from multiple images
+ */
+export class BatchImportManager {
+    constructor() {
+        this.initialized = false;
+        this.inputMode = 'urls'; // 'urls' or 'directory'
+        this.operationId = null;
+        this.wsConnection = null;
+        this.pollingInterval = null;
+        this.progress = null;
+        this.results = null;
+        this.isCancelled = false;
+    }
+
+    /**
+     * Show the batch import modal
+     */
+    showModal() {
+        if (!this.initialized) {
+            this.initialize();
+        }
+        this.resetState();
+        modalManager.showModal('batchImportModal');
+    }
+
+    /**
+     * Initialize the manager
+     */
+    initialize() {
+        this.initialized = true;
+
+        // Add event listener for persisting "Skip images without metadata" choice
+        const skipNoMetadata = document.getElementById('batchSkipNoMetadata');
+        if (skipNoMetadata) {
+            skipNoMetadata.addEventListener('change', (e) => {
+                setStorageItem('batch_import_skip_no_metadata', e.target.checked);
+            });
+        }
+    }
+
+    /**
+     * Reset all state to initial values
+     */
+    resetState() {
+        this.inputMode = 'urls';
+        this.operationId = null;
+        this.progress = null;
+        this.results = null;
+        this.isCancelled = false;
+
+        // Reset UI
+        this.showStep('batchInputStep');
+        this.toggleInputMode('urls');
+        
+        // Clear inputs
+        const urlInput = document.getElementById('batchUrlInput');
+        if (urlInput) urlInput.value = '';
+        
+        const directoryInput = document.getElementById('batchDirectoryInput');
+        if (directoryInput) directoryInput.value = '';
+        
+        const tagsInput = document.getElementById('batchTagsInput');
+        if (tagsInput) tagsInput.value = '';
+        
+        const skipNoMetadata = document.getElementById('batchSkipNoMetadata');
+        if (skipNoMetadata) {
+            // Load preference from storage, defaulting to true
+            skipNoMetadata.checked = getStorageItem('batch_import_skip_no_metadata', true);
+        }
+        
+        const recursiveCheck = document.getElementById('batchRecursiveCheck');
+        if (recursiveCheck) recursiveCheck.checked = true;
+
+        // Reset progress UI
+        this.updateProgressUI({
+            total: 0,
+            completed: 0,
+            success: 0,
+            failed: 0,
+            skipped: 0,
+            progress_percent: 0,
+            current_item: '',
+            status: 'pending'
+        });
+
+        // Reset results
+        const detailsList = document.getElementById('batchDetailsList');
+        if (detailsList) {
+            detailsList.innerHTML = '';
+            detailsList.style.display = 'none';
+        }
+
+        const toggleIcon = document.getElementById('resultsToggleIcon');
+        if (toggleIcon) {
+            toggleIcon.classList.remove('expanded');
+        }
+
+        // Clean up any existing connections
+        this.cleanupConnections();
+    }
+
+    /**
+     * Show a specific step in the modal
+     */
+    showStep(stepId) {
+        document.querySelectorAll('.batch-import-step').forEach(step => {
+            step.style.display = 'none';
+        });
+        
+        const step = document.getElementById(stepId);
+        if (step) {
+            step.style.display = 'block';
+        }
+    }
+
+    /**
+     * Toggle between URL list and directory input modes
+     */
+    toggleInputMode(mode) {
+        this.inputMode = mode;
+
+        // Update toggle buttons
+        document.querySelectorAll('.toggle-btn[data-mode]').forEach(btn => {
+            btn.classList.remove('active');
+        });
+        
+        const activeBtn = document.querySelector(`.toggle-btn[data-mode="${mode}"]`);
+        if (activeBtn) {
+            activeBtn.classList.add('active');
+        }
+
+        // Show/hide appropriate sections
+        const urlSection = document.getElementById('urlListSection');
+        const directorySection = document.getElementById('directorySection');
+
+        if (urlSection && directorySection) {
+            if (mode === 'urls') {
+                urlSection.style.display = 'block';
+                directorySection.style.display = 'none';
+            } else {
+                urlSection.style.display = 'none';
+                directorySection.style.display = 'block';
+            }
+        }
+    }
+
+    /**
+     * Start the batch import process
+     */
+    async startImport() {
+        const data = this.collectInputData();
+        
+        if (!this.validateInput(data)) {
+            return;
+        }
+
+        try {
+            // Show progress step
+            this.showStep('batchProgressStep');
+            
+            // Start the import
+            const response = await this.sendStartRequest(data);
+            
+            if (response.success) {
+                this.operationId = response.operation_id;
+                this.isCancelled = false;
+                
+                // Connect to WebSocket for real-time updates
+                this.connectWebSocket();
+                
+                // Start polling as fallback
+                this.startPolling();
+            } else {
+                showToast('toast.recipes.batchImportFailed', { message: response.error }, 'error');
+                this.showStep('batchInputStep');
+            }
+        } catch (error) {
+            console.error('Error starting batch import:', error);
+            showToast('toast.recipes.batchImportFailed', { message: error.message }, 'error');
+            this.showStep('batchInputStep');
+        }
+    }
+
+    /**
+     * Collect input data from the form
+     */
+    collectInputData() {
+        const data = {
+            mode: this.inputMode,
+            tags: [],
+            skip_no_metadata: false
+        };
+
+        // Collect tags
+        const tagsInput = document.getElementById('batchTagsInput');
+        if (tagsInput && tagsInput.value.trim()) {
+            data.tags = tagsInput.value.split(',').map(t => t.trim()).filter(t => t);
+        }
+
+        // Collect skip_no_metadata
+        const skipNoMetadata = document.getElementById('batchSkipNoMetadata');
+        if (skipNoMetadata) {
+            data.skip_no_metadata = skipNoMetadata.checked;
+        }
+
+        if (this.inputMode === 'urls') {
+            const urlInput = document.getElementById('batchUrlInput');
+            if (urlInput) {
+                const urls = urlInput.value.split('\n')
+                    .map(line => line.trim())
+                    .filter(line => line.length > 0);
+                
+                // Convert to items format
+                data.items = urls.map(url => ({
+                    source: url,
+                    type: this.detectUrlType(url)
+                }));
+            }
+        } else {
+            const directoryInput = document.getElementById('batchDirectoryInput');
+            if (directoryInput) {
+                data.directory = directoryInput.value.trim();
+            }
+            
+            const recursiveCheck = document.getElementById('batchRecursiveCheck');
+            if (recursiveCheck) {
+                data.recursive = recursiveCheck.checked;
+            }
+        }
+
+        return data;
+    }
+
+    /**
+     * Detect if a URL is http or local path
+     */
+    detectUrlType(url) {
+        if (url.startsWith('http://') || url.startsWith('https://')) {
+            return 'url';
+        }
+        return 'local_path';
+    }
+
+    /**
+     * Validate the input data
+     */
+    validateInput(data) {
+        if (data.mode === 'urls') {
+            if (!data.items || data.items.length === 0) {
+                showToast('toast.recipes.batchImportNoUrls', {}, 'error');
+                return false;
+            }
+        } else {
+            if (!data.directory) {
+                showToast('toast.recipes.batchImportNoDirectory', {}, 'error');
+                return false;
+            }
+        }
+        return true;
+    }
+
+    /**
+     * Send the start batch import request
+     */
+    async sendStartRequest(data) {
+        const endpoint = data.mode === 'urls' 
+            ? '/api/lm/recipes/batch-import/start'
+            : '/api/lm/recipes/batch-import/directory';
+
+        const response = await fetch(endpoint, {
+            method: 'POST',
+            headers: {
+                'Content-Type': 'application/json'
+            },
+            body: JSON.stringify(data)
+        });
+
+        return await response.json();
+    }
+
+    /**
+     * Connect to WebSocket for real-time progress updates
+     */
+    connectWebSocket() {
+        const wsProtocol = window.location.protocol === 'https:' ? 'wss:' : 'ws:';
+        const wsUrl = `${wsProtocol}//${window.location.host}/ws/batch-import-progress?id=${this.operationId}`;
+        
+        this.wsConnection = new WebSocket(wsUrl);
+        
+        this.wsConnection.onopen = () => {
+            console.log('Connected to batch import progress WebSocket');
+        };
+        
+        this.wsConnection.onmessage = (event) => {
+            try {
+                const data = JSON.parse(event.data);
+                if (data.type === 'batch_import_progress') {
+                    this.handleProgressUpdate(data);
+                }
+            } catch (error) {
+                console.error('Error parsing WebSocket message:', error);
+            }
+        };
+        
+        this.wsConnection.onerror = (error) => {
+            console.error('WebSocket error:', error);
+        };
+        
+        this.wsConnection.onclose = () => {
+            console.log('WebSocket connection closed');
+        };
+    }
+
+    /**
+     * Start polling for progress updates (fallback)
+     */
+    startPolling() {
+        this.pollingInterval = setInterval(async () => {
+            if (!this.operationId || this.isCancelled) {
+                return;
+            }
+            
+            try {
+                const response = await fetch(`/api/lm/recipes/batch-import/progress?operation_id=${this.operationId}`);
+                const data = await response.json();
+                
+                if (data.success && data.progress) {
+                    this.handleProgressUpdate(data.progress);
+                }
+            } catch (error) {
+                console.error('Error polling progress:', error);
+            }
+        }, 1000);
+    }
+
+    /**
+     * Handle progress update from WebSocket or polling
+     */
+    handleProgressUpdate(progress) {
+        this.progress = progress;
+        this.updateProgressUI(progress);
+        
+        // Check if import is complete
+        if (progress.status === 'completed' || progress.status === 'cancelled' || 
+            (progress.total > 0 && progress.completed >= progress.total)) {
+            this.importComplete(progress);
+        }
+    }
+
+    /**
+     * Update the progress UI
+     */
+    updateProgressUI(progress) {
+        // Update progress bar
+        const progressBar = document.getElementById('batchProgressBar');
+        if (progressBar) {
+            progressBar.style.width = `${progress.progress_percent || 0}%`;
+        }
+        
+        // Update percentage
+        const progressPercent = document.getElementById('batchProgressPercent');
+        if (progressPercent) {
+            progressPercent.textContent = `${Math.round(progress.progress_percent || 0)}%`;
+        }
+        
+        // Update stats
+        const totalCount = document.getElementById('batchTotalCount');
+        if (totalCount) totalCount.textContent = progress.total || 0;
+        
+        const successCount = document.getElementById('batchSuccessCount');
+        if (successCount) successCount.textContent = progress.success || 0;
+        
+        const failedCount = document.getElementById('batchFailedCount');
+        if (failedCount) failedCount.textContent = progress.failed || 0;
+        
+        const skippedCount = document.getElementById('batchSkippedCount');
+        if (skippedCount) skippedCount.textContent = progress.skipped || 0;
+        
+        // Update current item
+        const currentItem = document.getElementById('batchCurrentItem');
+        if (currentItem) {
+            currentItem.textContent = progress.current_item || '-';
+        }
+        
+        // Update status text
+        const statusText = document.getElementById('batchStatusText');
+        if (statusText) {
+            if (progress.status === 'running') {
+                statusText.textContent = translate('recipes.batchImport.importing', {}, 'Importing...');
+            } else if (progress.status === 'completed') {
+                statusText.textContent = translate('recipes.batchImport.completed', {}, 'Import completed');
+            } else if (progress.status === 'cancelled') {
+                statusText.textContent = translate('recipes.batchImport.cancelled', {}, 'Import cancelled');
+            }
+        }
+
+        // Update container classes
+        const progressContainer = document.querySelector('.batch-progress-container');
+        if (progressContainer) {
+            progressContainer.classList.remove('completed', 'cancelled', 'error');
+            if (progress.status === 'completed') {
+                progressContainer.classList.add('completed');
+            } else if (progress.status === 'cancelled') {
+                progressContainer.classList.add('cancelled');
+            } else if (progress.failed > 0 && progress.failed === progress.total) {
+                progressContainer.classList.add('error');
+            }
+        }
+    }
+
+    /**
+     * Handle import completion
+     */
+    importComplete(progress) {
+        this.cleanupConnections();
+        this.results = progress;
+        
+        // Refresh recipes list to show newly imported recipes
+        if (window.recipeManager && typeof window.recipeManager.loadRecipes === 'function') {
+            window.recipeManager.loadRecipes();
+        }
+        
+        // Show results step
+        this.showStep('batchResultsStep');
+        this.updateResultsUI(progress);
+    }
+
+    /**
+     * Update the results UI
+     */
+    updateResultsUI(progress) {
+        // Update summary cards
+        const resultsTotal = document.getElementById('resultsTotal');
+        if (resultsTotal) resultsTotal.textContent = progress.total || 0;
+        
+        const resultsSuccess = document.getElementById('resultsSuccess');
+        if (resultsSuccess) resultsSuccess.textContent = progress.success || 0;
+        
+        const resultsFailed = document.getElementById('resultsFailed');
+        if (resultsFailed) resultsFailed.textContent = progress.failed || 0;
+        
+        const resultsSkipped = document.getElementById('resultsSkipped');
+        if (resultsSkipped) resultsSkipped.textContent = progress.skipped || 0;
+        
+        // Update header based on results
+        const resultsHeader = document.getElementById('batchResultsHeader');
+        if (resultsHeader) {
+            const icon = resultsHeader.querySelector('.results-icon i');
+            const title = resultsHeader.querySelector('.results-title');
+            
+                if (this.isCancelled) {
+                if (icon) {
+                    icon.className = 'fas fa-stop-circle';
+                    icon.parentElement.classList.add('warning');
+                }
+                if (title) title.textContent = translate('recipes.batchImport.cancelled', {}, 'Import cancelled');
+            } else if (progress.failed === 0 && progress.success > 0) {
+                if (icon) {
+                    icon.className = 'fas fa-check-circle';
+                    icon.parentElement.classList.remove('warning', 'error');
+                }
+                if (title) title.textContent = translate('recipes.batchImport.completed', {}, 'Import completed');
+            } else if (progress.failed > 0 && progress.success === 0) {
+                if (icon) {
+                    icon.className = 'fas fa-times-circle';
+                    icon.parentElement.classList.add('error');
+                }
+                if (title) title.textContent = translate('recipes.batchImport.failed', {}, 'Import failed');
+            } else {
+                if (icon) {
+                    icon.className = 'fas fa-exclamation-circle';
+                    icon.parentElement.classList.add('warning');
+                }
+                if (title) title.textContent = translate('recipes.batchImport.completedWithErrors', {}, 'Completed with errors');
+            }
+        }
+    }
+
+    /**
+     * Toggle the results details visibility
+     */
+    toggleResultsDetails() {
+        const detailsList = document.getElementById('batchDetailsList');
+        const toggleIcon = document.getElementById('resultsToggleIcon');
+        const toggle = document.querySelector('.details-toggle');
+        
+        if (detailsList && toggleIcon) {
+            if (detailsList.style.display === 'none') {
+                detailsList.style.display = 'block';
+                toggleIcon.classList.add('expanded');
+                if (toggle) toggle.classList.add('expanded');
+                
+                // Load details if not loaded
+                if (detailsList.children.length === 0 && this.results && this.results.items) {
+                    this.loadResultsDetails(this.results.items);
+                }
+            } else {
+                detailsList.style.display = 'none';
+                toggleIcon.classList.remove('expanded');
+                if (toggle) toggle.classList.remove('expanded');
+            }
+        }
+    }
+
+    /**
+     * Load results details into the list
+     */
+    loadResultsDetails(items) {
+        const detailsList = document.getElementById('batchDetailsList');
+        if (!detailsList) return;
+        
+        detailsList.innerHTML = '';
+        
+        items.forEach(item => {
+            const resultItem = document.createElement('div');
+            resultItem.className = 'result-item';
+            
+            const statusClass = item.status === 'success' ? 'success' : 
+                               item.status === 'failed' ? 'failed' : 'skipped';
+            const statusIcon = item.status === 'success' ? 'check' : 
+                              item.status === 'failed' ? 'times' : 'forward';
+            
+            resultItem.innerHTML = `
+                <div class="result-item-status ${statusClass}">
+                    <i class="fas fa-${statusIcon}"></i>
+                </div>
+                <div class="result-item-info">
+                    <div class="result-item-name">${this.escapeHtml(item.source || item.current_item || 'Unknown')}</div>
+                    ${item.error_message ? `<div class="result-item-error">${this.escapeHtml(item.error_message)}</div>` : ''}
+                </div>
+            `;
+            
+            detailsList.appendChild(resultItem);
+        });
+    }
+
+    /**
+     * Cancel the current import
+     */
+    async cancelImport() {
+        if (!this.operationId) return;
+        
+        this.isCancelled = true;
+        
+        try {
+            const response = await fetch('/api/lm/recipes/batch-import/cancel', {
+                method: 'POST',
+                headers: {
+                    'Content-Type': 'application/json'
+                },
+                body: JSON.stringify({ operation_id: this.operationId })
+            });
+            
+            const data = await response.json();
+            
+            if (data.success) {
+                showToast('toast.recipes.batchImportCancelling', {}, 'info');
+            } else {
+                showToast('toast.recipes.batchImportCancelFailed', { message: data.error }, 'error');
+            }
+        } catch (error) {
+            console.error('Error cancelling import:', error);
+            showToast('toast.recipes.batchImportCancelFailed', { message: error.message }, 'error');
+        }
+    }
+
+    /**
+     * Close modal and reset state
+     */
+    closeAndReset() {
+        this.cleanupConnections();
+        this.resetState();
+        modalManager.closeModal('batchImportModal');
+    }
+
+    /**
+     * Start a new import (from results step)
+     */
+    startNewImport() {
+        this.resetState();
+        this.showStep('batchInputStep');
+    }
+
+    /**
+     * Toggle directory browser visibility
+     */
+    toggleDirectoryBrowser() {
+        const browser = document.getElementById('batchDirectoryBrowser');
+        if (browser) {
+            const isVisible = browser.style.display !== 'none';
+            browser.style.display = isVisible ? 'none' : 'block';
+            
+            if (!isVisible) {
+                // Load initial directory when opening
+                const currentPath = document.getElementById('batchDirectoryInput').value;
+                this.loadDirectory(currentPath || '/');
+            }
+        }
+    }
+
+    /**
+     * Load directory contents
+     */
+    async loadDirectory(path) {
+        try {
+            const response = await fetch('/api/lm/recipes/browse-directory', {
+                method: 'POST',
+                headers: {
+                    'Content-Type': 'application/json'
+                },
+                body: JSON.stringify({ path })
+            });
+
+            const data = await response.json();
+
+            if (data.success) {
+                this.renderDirectoryBrowser(data);
+            } else {
+                showToast('toast.recipes.batchImportBrowseFailed', { message: data.error }, 'error');
+            }
+        } catch (error) {
+            console.error('Error loading directory:', error);
+            showToast('toast.recipes.batchImportBrowseFailed', { message: error.message }, 'error');
+        }
+    }
+
+    /**
+     * Render directory browser UI
+     */
+    renderDirectoryBrowser(data) {
+        const currentPathEl = document.getElementById('batchCurrentPath');
+        const folderList = document.getElementById('batchFolderList');
+        const fileList = document.getElementById('batchFileList');
+        const directoryCount = document.getElementById('batchDirectoryCount');
+        const imageCount = document.getElementById('batchImageCount');
+
+        if (currentPathEl) {
+            currentPathEl.textContent = data.current_path;
+        }
+
+        // Render folders
+        if (folderList) {
+            folderList.innerHTML = '';
+            
+            // Add parent directory if available
+            if (data.parent_path) {
+                const parentItem = this.createFolderItem('..', data.parent_path, true);
+                folderList.appendChild(parentItem);
+            }
+
+            data.directories.forEach(dir => {
+                folderList.appendChild(this.createFolderItem(dir.name, dir.path));
+            });
+        }
+
+        // Render files
+        if (fileList) {
+            fileList.innerHTML = '';
+            data.image_files.forEach(file => {
+                fileList.appendChild(this.createFileItem(file.name, file.path, file.size));
+            });
+        }
+
+        // Update stats
+        if (directoryCount) {
+            directoryCount.textContent = data.directory_count;
+        }
+        if (imageCount) {
+            imageCount.textContent = data.image_count;
+        }
+    }
+
+    /**
+     * Create folder item element
+     */
+    createFolderItem(name, path, isParent = false) {
+        const item = document.createElement('div');
+        item.className = 'folder-item';
+        item.dataset.path = path;
+        
+        item.innerHTML = `
+            <i class="fas fa-folder${isParent ? '' : ''}"></i>
+            <span class="item-name">${this.escapeHtml(name)}</span>
+        `;
+        
+        item.addEventListener('click', () => {
+            if (isParent) {
+                this.navigateToParentDirectory();
+            } else {
+                this.loadDirectory(path);
+            }
+        });
+        
+        return item;
+    }
+
+    /**
+     * Create file item element
+     */
+    createFileItem(name, path, size) {
+        const item = document.createElement('div');
+        item.className = 'file-item';
+        item.dataset.path = path;
+        
+        item.innerHTML = `
+            <i class="fas fa-image"></i>
+            <span class="item-name">${this.escapeHtml(name)}</span>
+            <span class="item-size">${this.formatFileSize(size)}</span>
+        `;
+        
+        return item;
+    }
+
+    /**
+     * Navigate to parent directory
+     */
+    navigateToParentDirectory() {
+        const currentPath = document.getElementById('batchCurrentPath')?.textContent;
+        if (currentPath) {
+            // Get parent path using path manipulation
+            const lastSeparator = currentPath.lastIndexOf('/');
+            const parentPath = lastSeparator > 0 ? currentPath.substring(0, lastSeparator) : currentPath;
+            this.loadDirectory(parentPath);
+        }
+    }
+
+    /**
+     * Select current directory
+     */
+    selectCurrentDirectory() {
+        const currentPath = document.getElementById('batchCurrentPath')?.textContent;
+        const directoryInput = document.getElementById('batchDirectoryInput');
+        
+        if (currentPath && directoryInput) {
+            directoryInput.value = currentPath;
+            this.toggleDirectoryBrowser(); // Close browser
+            showToast('toast.recipes.batchImportDirectorySelected', { path: currentPath }, 'success');
+        }
+    }
+
+    /**
+     * Format file size for display
+     */
+    formatFileSize(bytes) {
+        if (bytes === 0) return '0 B';
+        const k = 1024;
+        const sizes = ['B', 'KB', 'MB', 'GB'];
+        const i = Math.floor(Math.log(bytes) / Math.log(k));
+        return Math.round(bytes / Math.pow(k, i) * 10) / 10 + ' ' + sizes[i];
+    }
+
+    /**
+     * Escape HTML to prevent XSS
+     */
+    escapeHtml(text) {
+        if (!text) return '';
+        const div = document.createElement('div');
+        div.textContent = text;
+        return div.innerHTML;
+    }
+
+    /**
+     * Browse for directory using File System Access API (deprecated - kept for compatibility)
+     */
+    async browseDirectory() {
+        // Now redirects to the new directory browser
+        this.toggleDirectoryBrowser();
+    }
+
+    /**
+     * Clean up WebSocket and polling connections
+     */
+    cleanupConnections() {
+        if (this.wsConnection) {
+            if (this.wsConnection.readyState === WebSocket.OPEN || 
+                this.wsConnection.readyState === WebSocket.CONNECTING) {
+                this.wsConnection.close();
+            }
+            this.wsConnection = null;
+        }
+        
+        if (this.pollingInterval) {
+            clearInterval(this.pollingInterval);
+            this.pollingInterval = null;
+        }
+    }
+
+    /**
+     * Escape HTML to prevent XSS
+     */
+    escapeHtml(text) {
+        if (!text) return '';
+        const div = document.createElement('div');
+        div.textContent = text;
+        return div.innerHTML;
+    }
+}
+
+// Create singleton instance
+export const batchImportManager = new BatchImportManager();
--- a/static/js/managers/BulkManager.js
+++ b/static/js/managers/BulkManager.js
@@ -568,7 +568,8 @@ export class BulkManager {
    }

    deselectItem(filepath) {
-        const card = document.querySelector(`.model-card[data-filepath="${filepath}"]`);
+        const escapedPath = this.escapeAttributeValue(filepath);
+        const card = document.querySelector(`.model-card[data-filepath="${escapedPath}"]`);
        if (card) {
            card.classList.remove('selected');
        }
@@ -632,7 +633,8 @@ export class BulkManager {
                for (const filepath of state.selectedModels) {
                    const metadata = metadataCache.get(filepath);
                    if (metadata) {
-                        const card = document.querySelector(`.model-card[data-filepath="${filepath}"]`);
+                        const escapedPath = this.escapeAttributeValue(filepath);
+                        const card = document.querySelector(`.model-card[data-filepath="${escapedPath}"]`);
                        if (card) {
                            this.updateMetadataCacheFromCard(filepath, card);
                        }
--- a/static/js/managers/ModalManager.js
+++ b/static/js/managers/ModalManager.js
@@ -134,6 +134,19 @@ export class ModalManager {
            });
        }

+        // Add batchImportModal registration
+        const batchImportModal = document.getElementById('batchImportModal');
+        if (batchImportModal) {
+            this.registerModal('batchImportModal', {
+                element: batchImportModal,
+                onClose: () => {
+                    this.getModal('batchImportModal').element.style.display = 'none';
+                    document.body.classList.remove('modal-open');
+                },
+                closeOnOutsideClick: true
+            });
+        }
+
        // Add recipeModal registration
        const recipeModal = document.getElementById('recipeModal');
        if (recipeModal) {
--- a/static/js/recipes.js
+++ b/static/js/recipes.js
@@ -1,13 +1,14 @@
 // Recipe manager module
 import { appCore } from './core.js';
 import { ImportManager } from './managers/ImportManager.js';
+import { BatchImportManager } from './managers/BatchImportManager.js';
 import { RecipeModal } from './components/RecipeModal.js';
 import { state, getCurrentPageState } from './state/index.js';
 import { getSessionItem, removeSessionItem } from './utils/storageHelpers.js';
 import { RecipeContextMenu } from './components/ContextMenu/index.js';
 import { DuplicatesManager } from './components/DuplicatesManager.js';
 import { refreshVirtualScroll } from './utils/infiniteScroll.js';
-import { refreshRecipes, RecipeSidebarApiClient } from './api/recipeApi.js';
+import { refreshRecipes, syncChanges, RecipeSidebarApiClient } from './api/recipeApi.js';
 import { sidebarManager } from './components/SidebarManager.js';

 class RecipePageControls {
@@ -27,7 +28,7 @@ class RecipePageControls {
            return;
        }

-        refreshVirtualScroll();
+        await syncChanges();
    }

    getSidebarApiClient() {
@@ -46,6 +47,10 @@ class RecipeManager {
        // Initialize ImportManager
        this.importManager = new ImportManager();

+        // Initialize BatchImportManager and make it globally accessible
+        this.batchImportManager = new BatchImportManager();
+        window.batchImportManager = this.batchImportManager;
+
        // Initialize RecipeModal
        this.recipeModal = new RecipeModal();

@@ -236,6 +241,70 @@ class RecipeManager {
                refreshVirtualScroll();
            });
        }
+
+        // Initialize dropdown functionality for refresh button
+        this.initDropdowns();
+    }
+
+    initDropdowns() {
+        // Handle dropdown toggles
+        const dropdownToggles = document.querySelectorAll('.dropdown-toggle');
+        dropdownToggles.forEach(toggle => {
+            toggle.addEventListener('click', (e) => {
+                e.stopPropagation();
+                const dropdownGroup = toggle.closest('.dropdown-group');
+                
+                // Close all other open dropdowns first
+                document.querySelectorAll('.dropdown-group.active').forEach(group => {
+                    if (group !== dropdownGroup) {
+                        group.classList.remove('active');
+                    }
+                });
+                
+                dropdownGroup.classList.toggle('active');
+            });
+        });
+
+        // Handle quick refresh option (Sync Changes)
+        const quickRefreshOption = document.querySelector('[data-action="quick-refresh"]');
+        if (quickRefreshOption) {
+            quickRefreshOption.addEventListener('click', (e) => {
+                e.stopPropagation();
+                this.pageControls.refreshModels(false);
+                this.closeDropdowns();
+            });
+        }
+
+        // Handle full rebuild option (Rebuild Cache)
+        const fullRebuildOption = document.querySelector('[data-action="full-rebuild"]');
+        if (fullRebuildOption) {
+            fullRebuildOption.addEventListener('click', (e) => {
+                e.stopPropagation();
+                this.pageControls.refreshModels(true);
+                this.closeDropdowns();
+            });
+        }
+
+        // Handle main refresh button (default: sync changes)
+        const refreshBtn = document.querySelector('[data-action="refresh"]');
+        if (refreshBtn) {
+            refreshBtn.addEventListener('click', () => {
+                this.pageControls.refreshModels(false);
+            });
+        }
+
+        // Close dropdowns when clicking outside
+        document.addEventListener('click', (e) => {
+            if (!e.target.closest('.dropdown-group')) {
+                this.closeDropdowns();
+            }
+        });
+    }
+
+    closeDropdowns() {
+        document.querySelectorAll('.dropdown-group.active').forEach(group => {
+            group.classList.remove('active');
+        });
    }

    // This method is kept for compatibility but now uses virtual scrolling
--- a/static/js/utils/modalUtils.js
+++ b/static/js/utils/modalUtils.js
@@ -7,7 +7,10 @@ let pendingExcludePath = null;
 export function showDeleteModal(filePath) {
    pendingDeletePath = filePath;
    
-    const card = document.querySelector(`.model-card[data-filepath="${filePath}"]`);
+    const escapedPath = window.CSS && typeof window.CSS.escape === 'function'
+        ? window.CSS.escape(filePath)
+        : filePath.replace(/["\\]/g, '\\$&');
+    const card = document.querySelector(`.model-card[data-filepath="${escapedPath}"]`);
    const modelName = card ? card.dataset.name : filePath.split('/').pop();
    const modal = modalManager.getModal('deleteModal').element;
    const modelInfo = modal.querySelector('.delete-model-info');
@@ -47,7 +50,10 @@ export function closeDeleteModal() {
 export function showExcludeModal(filePath) {
    pendingExcludePath = filePath;
    
-    const card = document.querySelector(`.model-card[data-filepath="${filePath}"]`);
+    const escapedPath = window.CSS && typeof window.CSS.escape === 'function'
+        ? window.CSS.escape(filePath)
+        : filePath.replace(/["\\]/g, '\\$&');
+    const card = document.querySelector(`.model-card[data-filepath="${escapedPath}"]`);
    const modelName = card ? card.dataset.name : filePath.split('/').pop();
    const modal = modalManager.getModal('excludeModal').element;
    const modelInfo = modal.querySelector('.exclude-model-info');
--- a/static/js/utils/uiHelpers.js
+++ b/static/js/utils/uiHelpers.js
@@ -197,7 +197,10 @@ export function openCivitaiByMetadata(civitaiId, versionId, modelName = null) {
 }

 export function openCivitai(filePath) {
-  const loraCard = document.querySelector(`.model-card[data-filepath="${filePath}"]`);
+  const escapedPath = window.CSS && typeof window.CSS.escape === 'function'
+    ? window.CSS.escape(filePath)
+    : filePath.replace(/["\\]/g, '\\$&');
+  const loraCard = document.querySelector(`.model-card[data-filepath="${escapedPath}"]`);
  if (!loraCard) return;

  const metaData = JSON.parse(loraCard.dataset.meta);
@@ -483,8 +486,12 @@ async function ensureRelativeModelPath(modelPath, collectionType) {
    return modelPath;
  }

+  // Remove model file extension (.safetensors, .ckpt, .pt, .bin) for cleaner matching
+  // Backend removes extensions from paths before matching, so search term should not include extension
+  const searchTerm = fileName.replace(/\.(safetensors|ckpt|pt|bin)$/i, '');
+
  try {
-    const response = await fetch(`/api/lm/${collectionType}/relative-paths?search=${encodeURIComponent(fileName)}&limit=10`);
+    const response = await fetch(`/api/lm/${collectionType}/relative-paths?search=${encodeURIComponent(searchTerm)}&limit=10`);
    if (!response.ok) {
      return modelPath;
    }
--- a/templates/components/batch_import_modal.html
+++ b/templates/components/batch_import_modal.html
@@ -0,0 +1,206 @@
+<div id="batchImportModal" class="modal">
+    <div class="modal-content">
+        <div class="modal-header">
+            <button class="close" onclick="modalManager.closeModal('batchImportModal')">&times;</button>
+            <h2>{{ t('recipes.batchImport.title') }}</h2>
+        </div>
+        
+        <!-- Step 1: Input Selection -->
+        <div class="batch-import-step" id="batchInputStep">
+            <div class="import-mode-toggle">
+                <button class="toggle-btn active" data-mode="urls" onclick="batchImportManager.toggleInputMode('urls')">
+                    <i class="fas fa-link"></i> {{ t('recipes.batchImport.urlList') }}
+                </button>
+                <button class="toggle-btn" data-mode="directory" onclick="batchImportManager.toggleInputMode('directory')">
+                    <i class="fas fa-folder"></i> {{ t('recipes.batchImport.directory') }}
+                </button>
+            </div>
+            
+            <!-- URL List Section -->
+            <div class="import-section" id="urlListSection">
+                <p class="section-description">{{ t('recipes.batchImport.urlDescription') }}</p>
+                <div class="input-group">
+                    <label for="batchUrlInput">{{ t('recipes.batchImport.urlsLabel') }}</label>
+                    <textarea id="batchUrlInput" rows="8" placeholder="{{ t('recipes.batchImport.urlsPlaceholder') }}"></textarea>
+                    <div class="input-hint">
+                        <i class="fas fa-info-circle"></i>
+                        {{ t('recipes.batchImport.urlsHint') }}
+                    </div>
+                </div>
+            </div>
+            
+            <!-- Directory Section -->
+            <div class="import-section" id="directorySection" style="display: none;">
+                <p class="section-description">{{ t('recipes.batchImport.directoryDescription') }}</p>
+                <div class="input-group">
+                    <label for="batchDirectoryInput">{{ t('recipes.batchImport.directoryPath') }}</label>
+                    <div class="input-with-button">
+                        <input type="text" id="batchDirectoryInput" placeholder="{{ t('recipes.batchImport.directoryPlaceholder') }}" autocomplete="off">
+                        <button class="secondary-btn" onclick="batchImportManager.toggleDirectoryBrowser()">
+                            <i class="fas fa-folder-open"></i> {{ t('recipes.batchImport.browse') }}
+                        </button>
+                    </div>
+                </div>
+                
+                <!-- Directory Browser -->
+                <div class="directory-browser" id="batchDirectoryBrowser" style="display: none;">
+                    <div class="browser-header">
+                        <button class="back-btn" onclick="batchImportManager.navigateToParentDirectory()" title="{{ t('recipes.batchImport.backToParent') }}">
+                            <i class="fas fa-arrow-up"></i>
+                        </button>
+                        <div class="current-path" id="batchCurrentPath"></div>
+                    </div>
+                    <div class="browser-content">
+                        <div class="browser-section">
+                            <div class="section-label"><i class="fas fa-folder"></i> {{ t('recipes.batchImport.folders') }}</div>
+                            <div class="folder-list" id="batchFolderList"></div>
+                        </div>
+                        <div class="browser-section">
+                            <div class="section-label"><i class="fas fa-image"></i> {{ t('recipes.batchImport.imageFiles') }}</div>
+                            <div class="file-list" id="batchFileList"></div>
+                        </div>
+                    </div>
+                    <div class="browser-footer">
+                        <div class="stats">
+                            <span id="batchDirectoryCount">0</span> {{ t('recipes.batchImport.folders') }}, 
+                            <span id="batchImageCount">0</span> {{ t('recipes.batchImport.images') }}
+                        </div>
+                        <button class="primary-btn" onclick="batchImportManager.selectCurrentDirectory()">
+                            <i class="fas fa-check"></i> {{ t('recipes.batchImport.selectFolder') }}
+                        </button>
+                    </div>
+                </div>
+                
+                <div class="checkbox-group">
+                    <label class="checkbox-label">
+                        <input type="checkbox" id="batchRecursiveCheck" checked>
+                        <span class="checkmark"></span>
+                        {{ t('recipes.batchImport.recursive') }}
+                    </label>
+                </div>
+            </div>
+            
+            <!-- Common Options -->
+            <div class="batch-options">
+                <div class="input-group">
+                    <label for="batchTagsInput">{{ t('recipes.batchImport.tagsOptional') }}</label>
+                    <input type="text" id="batchTagsInput" placeholder="{{ t('recipes.batchImport.tagsPlaceholder') }}">
+                    <div class="input-hint">
+                        <i class="fas fa-info-circle"></i>
+                        {{ t('recipes.batchImport.tagsHint') }}
+                    </div>
+                </div>
+                
+                <div class="checkbox-group">
+                    <label class="checkbox-label">
+                        <input type="checkbox" id="batchSkipNoMetadata">
+                        <span class="checkmark"></span>
+                        {{ t('recipes.batchImport.skipNoMetadata') }}
+                    </label>
+                </div>
+            </div>
+            
+            <div class="modal-actions">
+                <button class="secondary-btn" onclick="modalManager.closeModal('batchImportModal')">{{ t('common.actions.cancel') }}</button>
+                <button class="primary-btn" id="batchImportStartBtn" onclick="batchImportManager.startImport()">
+                    <i class="fas fa-play"></i> {{ t('recipes.batchImport.start') }}
+                </button>
+            </div>
+        </div>
+        
+        <!-- Step 2: Progress -->
+        <div class="batch-import-step" id="batchProgressStep" style="display: none;">
+            <div class="batch-progress-container">
+                <div class="progress-header">
+                    <div class="progress-status">
+                        <span class="status-icon"><i class="fas fa-spinner fa-spin"></i></span>
+                        <span class="status-text" id="batchStatusText">{{ t('recipes.batchImport.importing') }}</span>
+                    </div>
+                    <div class="progress-percentage" id="batchProgressPercent">0%</div>
+                </div>
+                
+                <div class="progress-bar-container">
+                    <div class="progress-bar" id="batchProgressBar" style="width: 0%"></div>
+                </div>
+                
+                <div class="progress-stats">
+                    <div class="stat-item">
+                        <span class="stat-label">{{ t('recipes.batchImport.total') }}</span>
+                        <span class="stat-value" id="batchTotalCount">0</span>
+                    </div>
+                    <div class="stat-item success">
+                        <span class="stat-label">{{ t('recipes.batchImport.success') }}</span>
+                        <span class="stat-value" id="batchSuccessCount">0</span>
+                    </div>
+                    <div class="stat-item failed">
+                        <span class="stat-label">{{ t('recipes.batchImport.failed') }}</span>
+                        <span class="stat-value" id="batchFailedCount">0</span>
+                    </div>
+                    <div class="stat-item skipped">
+                        <span class="stat-label">{{ t('recipes.batchImport.skipped') }}</span>
+                        <span class="stat-value" id="batchSkippedCount">0</span>
+                    </div>
+                </div>
+                
+                <div class="current-item" id="batchCurrentItemContainer">
+                    <span class="current-item-label">{{ t('recipes.batchImport.current') }}</span>
+                    <span class="current-item-name" id="batchCurrentItem">-</span>
+                </div>
+            </div>
+            
+            <div class="modal-actions">
+                <button class="secondary-btn" id="batchCancelBtn" onclick="batchImportManager.cancelImport()">
+                    <i class="fas fa-stop"></i> {{ t('recipes.batchImport.cancel') }}
+                </button>
+            </div>
+        </div>
+        
+        <!-- Step 3: Results -->
+        <div class="batch-import-step" id="batchResultsStep" style="display: none;">
+            <div class="batch-results-container">
+                <div class="results-header" id="batchResultsHeader">
+                    <div class="results-icon">
+                        <i class="fas fa-check-circle"></i>
+                    </div>
+                    <div class="results-title">{{ t('recipes.batchImport.completed') }}</div>
+                </div>
+                
+                <div class="results-summary">
+                    <div class="result-card total">
+                        <span class="result-label">{{ t('recipes.batchImport.total') }}</span>
+                        <span class="result-value" id="resultsTotal">0</span>
+                    </div>
+                    <div class="result-card success">
+                        <span class="result-label">{{ t('recipes.batchImport.success') }}</span>
+                        <span class="result-value" id="resultsSuccess">0</span>
+                    </div>
+                    <div class="result-card failed">
+                        <span class="result-label">{{ t('recipes.batchImport.failed') }}</span>
+                        <span class="result-value" id="resultsFailed">0</span>
+                    </div>
+                    <div class="result-card skipped">
+                        <span class="result-label">{{ t('recipes.batchImport.skipped') }}</span>
+                        <span class="result-value" id="resultsSkipped">0</span>
+                    </div>
+                </div>
+                
+                <div class="results-details" id="batchResultsDetails">
+                    <div class="details-toggle" onclick="batchImportManager.toggleResultsDetails()">
+                        <i class="fas fa-chevron-down" id="resultsToggleIcon"></i>
+                        <span>{{ t('recipes.batchImport.viewDetails') }}</span>
+                    </div>
+                    <div class="details-list" id="batchDetailsList" style="display: none;">
+                        <!-- Details will be populated dynamically -->
+                    </div>
+                </div>
+            </div>
+            
+            <div class="modal-actions">
+                <button class="secondary-btn" onclick="batchImportManager.closeAndReset()">{{ t('common.actions.close') }}</button>
+                <button class="primary-btn" onclick="batchImportManager.startNewImport()">
+                    <i class="fas fa-plus"></i> {{ t('recipes.batchImport.newImport') }}
+                </button>
+            </div>
+        </div>
+    </div>
+</div>
--- a/templates/recipes.html
+++ b/templates/recipes.html
@@ -7,10 +7,12 @@
 <link rel="stylesheet" href="/loras_static/css/components/card.css?v={{ version }}">
 <link rel="stylesheet" href="/loras_static/css/components/recipe-modal.css?v={{ version }}">
 <link rel="stylesheet" href="/loras_static/css/components/import-modal.css?v={{ version }}">
+<link rel="stylesheet" href="/loras_static/css/components/batch-import-modal.css?v={{ version }}">
 {% endblock %}

 {% block additional_components %}
 {% include 'components/import_modal.html' %}
+{% include 'components/batch_import_modal.html' %}
 {% include 'components/recipe_modal.html' %}

 <div id="recipeContextMenu" class="context-menu" style="display: none;">
@@ -66,15 +68,29 @@
                    </optgroup>
                </select>
            </div>
-            <div title="{{ t('recipes.controls.refresh.title') }}" class="control-group">
-                <button onclick="recipeManager.refreshRecipes()"><i class="fas fa-sync"></i> {{
-                    t('common.actions.refresh')
-                    }}</button>
+            <div title="{{ t('recipes.controls.refresh.title') }}" class="control-group dropdown-group">
+                <button data-action="refresh" class="dropdown-main"><i class="fas fa-sync"></i> <span>{{
+                    t('common.actions.refresh') }}</span></button>
+                <button class="dropdown-toggle" aria-label="Show refresh options">
+                    <i class="fas fa-caret-down"></i>
+                </button>
+                <div class="dropdown-menu">
+                    <div class="dropdown-item" data-action="quick-refresh" title="{{ t('recipes.controls.refresh.quickTooltip', default='Sync changes - quick refresh without rebuilding cache') }}">
+                        <i class="fas fa-bolt"></i> <span>{{ t('loras.controls.refresh.quick', default='Sync Changes') }}</span>
+                    </div>
+                    <div class="dropdown-item" data-action="full-rebuild" title="{{ t('recipes.controls.refresh.fullTooltip', default='Rebuild cache - full rescan of all recipe files') }}">
+                        <i class="fas fa-tools"></i> <span>{{ t('loras.controls.refresh.full', default='Rebuild Cache') }}</span>
+                    </div>
+                </div>
            </div>
            <div title="{{ t('recipes.controls.import.title') }}" class="control-group">
                <button onclick="importManager.showImportModal()"><i class="fas fa-file-import"></i> {{
                    t('recipes.controls.import.action') }}</button>
            </div>
+            <div title="{{ t('recipes.batchImport.title') }}" class="control-group">
+                <button onclick="batchImportManager.showModal()"><i class="fas fa-layer-group"></i> {{
+                    t('recipes.batchImport.action') }}</button>
+            </div>
            <div class="control-group" title="{{ t('loras.controls.bulk.title') }}">
                <button id="bulkOperationsBtn" data-action="bulk" title="{{ t('loras.controls.bulk.title') }}">
                    <i class="fas fa-th-large"></i> <span><span>{{ t('loras.controls.bulk.action') }}</span>
--- a/tests/config/test_checkpoint_path_overlap.py
+++ b/tests/config/test_checkpoint_path_overlap.py
@@ -36,8 +36,8 @@ class TestCheckpointPathOverlap:
            config._preview_root_paths = set()
            config._cached_fingerprint = None

-            # Call the method under test
-            result = config._prepare_checkpoint_paths(
+            # Call the method under test - now returns a tuple
+            all_paths, checkpoint_roots, unet_roots = config._prepare_checkpoint_paths(
                [str(checkpoints_link)], [str(unet_link)]
            )

@@ -50,21 +50,27 @@ class TestCheckpointPathOverlap:
        ]
        assert len(warning_messages) == 1
        assert "checkpoints" in warning_messages[0].lower()
-        assert "diffusion_models" in warning_messages[0].lower() or "unet" in warning_messages[0].lower()
+        assert (
+            "diffusion_models" in warning_messages[0].lower()
+            or "unet" in warning_messages[0].lower()
+        )
        # Verify warning mentions backward compatibility fallback
-        assert "falling back" in warning_messages[0].lower() or "backward compatibility" in warning_messages[0].lower()
+        assert (
+            "falling back" in warning_messages[0].lower()
+            or "backward compatibility" in warning_messages[0].lower()
+        )

        # Verify only one path is returned (deduplication still works)
-        assert len(result) == 1
+        assert len(all_paths) == 1
        # Prioritizes checkpoints path for backward compatibility
-        assert _normalize(result[0]) == _normalize(str(checkpoints_link))
+        assert _normalize(all_paths[0]) == _normalize(str(checkpoints_link))

-        # Verify checkpoints_roots has the path (prioritized)
-        assert len(config.checkpoints_roots) == 1
-        assert _normalize(config.checkpoints_roots[0]) == _normalize(str(checkpoints_link))
+        # Verify checkpoint_roots has the path (prioritized)
+        assert len(checkpoint_roots) == 1
+        assert _normalize(checkpoint_roots[0]) == _normalize(str(checkpoints_link))

        # Verify unet_roots is empty (overlapping paths removed)
-        assert config.unet_roots == []
+        assert unet_roots == []

    def test_non_overlapping_paths_no_warning(
        self, monkeypatch: pytest.MonkeyPatch, tmp_path, caplog
@@ -83,7 +89,7 @@ class TestCheckpointPathOverlap:
            config._preview_root_paths = set()
            config._cached_fingerprint = None

-            result = config._prepare_checkpoint_paths(
+            all_paths, checkpoint_roots, unet_roots = config._prepare_checkpoint_paths(
                [str(checkpoints_dir)], [str(unet_dir)]
            )

@@ -97,14 +103,14 @@ class TestCheckpointPathOverlap:
        assert len(warning_messages) == 0

        # Verify both paths are returned
-        assert len(result) == 2
-        normalized_result = [_normalize(p) for p in result]
+        assert len(all_paths) == 2
+        normalized_result = [_normalize(p) for p in all_paths]
        assert _normalize(str(checkpoints_dir)) in normalized_result
        assert _normalize(str(unet_dir)) in normalized_result

        # Verify both roots are properly set
-        assert len(config.checkpoints_roots) == 1
-        assert len(config.unet_roots) == 1
+        assert len(checkpoint_roots) == 1
+        assert len(unet_roots) == 1

    def test_partial_overlap_prioritizes_checkpoints(
        self, monkeypatch: pytest.MonkeyPatch, tmp_path, caplog
@@ -129,9 +135,9 @@ class TestCheckpointPathOverlap:
            config._cached_fingerprint = None

            # One checkpoint path overlaps with one unet path
-            result = config._prepare_checkpoint_paths(
+            all_paths, checkpoint_roots, unet_roots = config._prepare_checkpoint_paths(
                [str(shared_link), str(separate_checkpoint)],
-                [str(shared_link), str(separate_unet)]
+                [str(shared_link), str(separate_unet)],
            )

        # Verify warning was logged for the overlapping path
@@ -144,17 +150,20 @@ class TestCheckpointPathOverlap:
        assert len(warning_messages) == 1

        # Verify 3 unique paths (shared counted once as checkpoint, plus separate ones)
-        assert len(result) == 3
+        assert len(all_paths) == 3

        # Verify the overlapping path appears in warning message
-        assert str(shared_link.name) in warning_messages[0] or str(shared_dir.name) in warning_messages[0]
+        assert (
+            str(shared_link.name) in warning_messages[0]
+            or str(shared_dir.name) in warning_messages[0]
+        )

-        # Verify checkpoints_roots includes both checkpoint paths (including the shared one)
-        assert len(config.checkpoints_roots) == 2
-        checkpoint_normalized = [_normalize(p) for p in config.checkpoints_roots]
+        # Verify checkpoint_roots includes both checkpoint paths (including the shared one)
+        assert len(checkpoint_roots) == 2
+        checkpoint_normalized = [_normalize(p) for p in checkpoint_roots]
        assert _normalize(str(shared_link)) in checkpoint_normalized
        assert _normalize(str(separate_checkpoint)) in checkpoint_normalized

        # Verify unet_roots only includes the non-overlapping unet path
-        assert len(config.unet_roots) == 1
-        assert _normalize(config.unet_roots[0]) == _normalize(str(separate_unet))
+        assert len(unet_roots) == 1
+        assert _normalize(unet_roots[0]) == _normalize(str(separate_unet))
--- a/tests/frontend/components/autocomplete.behavior.test.js
+++ b/tests/frontend/components/autocomplete.behavior.test.js
@@ -90,7 +90,7 @@ describe('AutoComplete widget interactions', () => {
    await vi.runAllTimersAsync();
    await Promise.resolve();

-    expect(fetchApiMock).toHaveBeenCalledWith('/lm/loras/relative-paths?search=example&limit=20');
+    expect(fetchApiMock).toHaveBeenCalledWith('/lm/loras/relative-paths?search=example&limit=100');
    const items = autoComplete.dropdown.querySelectorAll('.comfy-autocomplete-item');
    expect(items).toHaveLength(1);
    expect(autoComplete.dropdown.style.display).toBe('block');
@@ -156,4 +156,542 @@ describe('AutoComplete widget interactions', () => {
    expect(highlighted).toContain('detail');
    expect(highlighted).not.toMatch(/beta<\/span>/i);
  });
+
+  it('handles arrow key navigation with virtual scrolling', async () => {
+    vi.useFakeTimers();
+
+    const mockItems = Array.from({ length: 50 }, (_, i) => `model_${i.toString().padStart(2, '0')}.safetensors`);
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, relative_paths: mockItems }),
+    });
+
+    caretHelperInstance.getBeforeCursor.mockReturnValue('model');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'loras', {
+      debounceDelay: 0,
+      showPreview: false,
+      enableVirtualScroll: true,
+      itemHeight: 40,
+      visibleItems: 15,
+      pageSize: 20,
+    });
+
+    input.value = 'model';
+    input.dispatchEvent(new Event('input', { bubbles: true }));
+
+    await vi.runAllTimersAsync();
+    await Promise.resolve();
+
+    expect(autoComplete.items.length).toBeGreaterThan(0);
+    expect(autoComplete.selectedIndex).toBe(0);
+
+    const initialSelectedEl = autoComplete.contentContainer?.querySelector('.comfy-autocomplete-item-selected');
+    expect(initialSelectedEl).toBeDefined();
+
+    const arrowDownEvent = new KeyboardEvent('keydown', { key: 'ArrowDown', bubbles: true });
+    input.dispatchEvent(arrowDownEvent);
+
+    expect(autoComplete.selectedIndex).toBe(1);
+
+    const secondSelectedEl = autoComplete.contentContainer?.querySelector('.comfy-autocomplete-item-selected');
+    expect(secondSelectedEl).toBeDefined();
+    expect(secondSelectedEl?.dataset.index).toBe('1');
+
+    const arrowUpEvent = new KeyboardEvent('keydown', { key: 'ArrowUp', bubbles: true });
+    input.dispatchEvent(arrowUpEvent);
+
+    expect(autoComplete.selectedIndex).toBe(0);
+
+    const firstSelectedElAgain = autoComplete.contentContainer?.querySelector('.comfy-autocomplete-item-selected');
+    expect(firstSelectedElAgain).toBeDefined();
+    expect(firstSelectedElAgain?.dataset.index).toBe('0');
+  });
+
+  it('maintains selection when scrolling to invisible items', async () => {
+    vi.useFakeTimers();
+
+    const mockItems = Array.from({ length: 100 }, (_, i) => `item_${i.toString().padStart(3, '0')}.safetensors`);
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, relative_paths: mockItems }),
+    });
+
+    caretHelperInstance.getBeforeCursor.mockReturnValue('item');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.style.width = '400px';
+    input.style.height = '200px';
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'loras', {
+      debounceDelay: 0,
+      showPreview: false,
+      enableVirtualScroll: true,
+      itemHeight: 40,
+      visibleItems: 15,
+      pageSize: 20,
+    });
+
+    input.value = 'item';
+    input.dispatchEvent(new Event('input', { bubbles: true }));
+
+    await vi.runAllTimersAsync();
+    await Promise.resolve();
+
+    expect(autoComplete.items.length).toBeGreaterThan(0);
+
+    autoComplete.selectedIndex = 14;
+
+    const scrollTopBefore = autoComplete.scrollContainer?.scrollTop || 0;
+
+    const arrowDownEvent = new KeyboardEvent('keydown', { key: 'ArrowDown', bubbles: true });
+    input.dispatchEvent(arrowDownEvent);
+
+    await vi.runAllTimersAsync();
+    await Promise.resolve();
+
+    expect(autoComplete.selectedIndex).toBe(15);
+
+    const selectedEl = autoComplete.contentContainer?.querySelector('.comfy-autocomplete-item-selected');
+    expect(selectedEl).toBeDefined();
+    expect(selectedEl?.dataset.index).toBe('15');
+
+    const scrollTopAfter = autoComplete.scrollContainer?.scrollTop || 0;
+    expect(scrollTopAfter).toBeGreaterThanOrEqual(scrollTopBefore);
+  });
+
+  it('replaces entire multi-word phrase when it matches selected tag (Danbooru convention)', async () => {
+    const mockTags = [
+      { tag_name: 'looking_to_the_side', category: 0, post_count: 1234 },
+      { tag_name: 'looking_away', category: 0, post_count: 5678 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    caretHelperInstance.getBeforeCursor.mockReturnValue('looking to the side');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'looking to the side';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+
+    await autoComplete.insertSelection('looking_to_the_side');
+
+    expect(input.value).toBe('looking_to_the_side, ');
+    expect(autoComplete.dropdown.style.display).toBe('none');
+    expect(input.focus).toHaveBeenCalled();
+  });
+
+  it('replaces only last token when typing partial match (e.g., "hello 1gi" -> "1girl")', async () => {
+    const mockTags = [
+      { tag_name: '1girl', category: 4, post_count: 500000 },
+      { tag_name: '1boy', category: 4, post_count: 300000 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    caretHelperInstance.getBeforeCursor.mockReturnValue('hello 1gi');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'hello 1gi';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = 'hello 1gi';
+
+    await autoComplete.insertSelection('1girl');
+
+    expect(input.value).toBe('hello 1girl, ');
+  });
+
+  it('replaces entire phrase for underscore tag match (e.g., "blue hair" -> "blue_hair")', async () => {
+    const mockTags = [
+      { tag_name: 'blue_hair', category: 0, post_count: 45000 },
+      { tag_name: 'blue_eyes', category: 0, post_count: 80000 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    caretHelperInstance.getBeforeCursor.mockReturnValue('blue hair');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'blue hair';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = 'blue hair';
+
+    await autoComplete.insertSelection('blue_hair');
+
+    expect(input.value).toBe('blue_hair, ');
+  });
+
+  it('handles multi-word phrase with preceding text correctly', async () => {
+    const mockTags = [
+      { tag_name: 'looking_to_the_side', category: 0, post_count: 1234 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    caretHelperInstance.getBeforeCursor.mockReturnValue('1girl, looking to the side');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = '1girl, looking to the side';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = 'looking to the side';
+
+    await autoComplete.insertSelection('looking_to_the_side');
+
+    expect(input.value).toBe('1girl, looking_to_the_side, ');
+  });
+
+  it('replaces entire command and search term when using command mode with multi-word phrase', async () => {
+    const mockTags = [
+      { tag_name: 'looking_to_the_side', category: 4, post_count: 1234 },
+      { tag_name: 'looking_away', category: 4, post_count: 5678 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    // Simulate "/char looking to the side" input
+    caretHelperInstance.getBeforeCursor.mockReturnValue('/char looking to the side');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = '/char looking to the side';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    // Set up command mode state
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = { categories: [4, 11], label: 'Character' };
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = '/char looking to the side';
+
+    await autoComplete.insertSelection('looking_to_the_side');
+
+    // Command part should be replaced along with search term
+    expect(input.value).toBe('looking_to_the_side, ');
+  });
+
+  it('replaces only last token when multi-word query does not exactly match selected tag', async () => {
+    const mockTags = [
+      { tag_name: 'blue_hair', category: 0, post_count: 45000 },
+      { tag_name: 'blue_eyes', category: 0, post_count: 80000 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    // User types "looking to the blue" but selects "blue_hair" (doesn't match entire phrase)
+    caretHelperInstance.getBeforeCursor.mockReturnValue('looking to the blue');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'looking to the blue';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = 'looking to the blue';
+
+    await autoComplete.insertSelection('blue_hair');
+
+    // Only "blue" should be replaced, not the entire phrase
+    expect(input.value).toBe('looking to the blue_hair, ');
+  });
+
+  it('handles multiple consecutive spaces in multi-word phrase correctly', async () => {
+    const mockTags = [
+      { tag_name: 'looking_to_the_side', category: 0, post_count: 1234 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    // Input with multiple spaces between words
+    caretHelperInstance.getBeforeCursor.mockReturnValue('looking  to   the side');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'looking  to   the side';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = 'looking  to   the side';
+
+    await autoComplete.insertSelection('looking_to_the_side');
+
+    // Multiple spaces should be normalized to single underscores for matching
+    expect(input.value).toBe('looking_to_the_side, ');
+  });
+
+  it('handles command mode with partial match replacing only last token', async () => {
+    const mockTags = [
+      { tag_name: 'blue_hair', category: 0, post_count: 45000 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    // Command mode but selected tag doesn't match entire search phrase
+    caretHelperInstance.getBeforeCursor.mockReturnValue('/general looking to the blue');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = '/general looking to the blue';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    // Command mode with activeCommand
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = { categories: [0, 7], label: 'General' };
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = '/general looking to the blue';
+
+    await autoComplete.insertSelection('blue_hair');
+
+    // In command mode, the entire command + search term should be replaced
+    expect(input.value).toBe('blue_hair, ');
+  });
+
+  it('replaces entire phrase when selected tag starts with underscore version of search term (prefix match)', async () => {
+    const mockTags = [
+      { tag_name: 'looking_to_the_side', category: 0, post_count: 1234 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    // User types partial phrase "looking to the" and selects "looking_to_the_side"
+    caretHelperInstance.getBeforeCursor.mockReturnValue('looking to the');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'looking to the';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = 'looking to the';
+
+    await autoComplete.insertSelection('looking_to_the_side');
+
+    // Entire phrase should be replaced with selected tag (with underscores)
+    expect(input.value).toBe('looking_to_the_side, ');
+  });
+
+  it('inserts tag with underscores regardless of space replacement setting', async () => {
+    const mockTags = [
+      { tag_name: 'blue_hair', category: 0, post_count: 45000 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    caretHelperInstance.getBeforeCursor.mockReturnValue('blue');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'blue';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+
+    await autoComplete.insertSelection('blue_hair');
+
+    // Tag should be inserted with underscores, not spaces
+    expect(input.value).toBe('blue_hair, ');
+  });
+
+  it('replaces entire phrase when selected tag ends with underscore version of search term (suffix match)', async () => {
+    const mockTags = [
+      { tag_name: 'looking_to_the_side', category: 0, post_count: 1234 },
+    ];
+
+    fetchApiMock.mockResolvedValue({
+      json: () => Promise.resolve({ success: true, words: mockTags }),
+    });
+
+    // User types suffix "to the side" and selects "looking_to_the_side"
+    caretHelperInstance.getBeforeCursor.mockReturnValue('to the side');
+    caretHelperInstance.getCursorOffset.mockReturnValue({ left: 15, top: 25 });
+
+    const input = document.createElement('textarea');
+    input.value = 'to the side';
+    input.selectionStart = input.value.length;
+    input.focus = vi.fn();
+    input.setSelectionRange = vi.fn();
+    document.body.append(input);
+
+    const { AutoComplete } = await import(AUTOCOMPLETE_MODULE);
+    const autoComplete = new AutoComplete(input, 'prompt', {
+      debounceDelay: 0,
+      showPreview: false,
+      minChars: 1,
+    });
+
+    autoComplete.searchType = 'custom_words';
+    autoComplete.activeCommand = null;
+    autoComplete.items = mockTags;
+    autoComplete.selectedIndex = 0;
+    autoComplete.currentSearchTerm = 'to the side';
+
+    await autoComplete.insertSelection('looking_to_the_side');
+
+    // Entire phrase should be replaced with selected tag
+    expect(input.value).toBe('looking_to_the_side, ');
+  });
 });
--- a/tests/frontend/versionDetection.test.js
+++ b/tests/frontend/versionDetection.test.js
@@ -0,0 +1,75 @@
+import { describe, it, expect } from 'vitest';
+
+describe('Version Detection Logic', () => {
+    const parseVersion = (versionStr) => {
+        if (!versionStr || typeof versionStr !== 'string') {
+            return [0, 0, 0];
+        }
+        
+        const cleanVersion = versionStr.replace(/^[vV]/, '').split('-')[0];
+        const parts = cleanVersion.split('.').map(part => parseInt(part, 10) || 0);
+        
+        while (parts.length < 3) {
+            parts.push(0);
+        }
+        
+        return parts;
+    };
+    
+    const compareVersions = (version1, version2) => {
+        const v1 = typeof version1 === 'string' ? parseVersion(version1) : version1;
+        const v2 = typeof version2 === 'string' ? parseVersion(version2) : version2;
+        
+        for (let i = 0; i < 3; i++) {
+            if (v1[i] > v2[i]) return 1;
+            if (v1[i] < v2[i]) return -1;
+        }
+        
+        return 0;
+    };
+    
+    const MIN_VERSION_FOR_ACTION_BAR = [1, 33, 9];
+    
+    const supportsActionBarButtons = (version) => {
+        return compareVersions(version, MIN_VERSION_FOR_ACTION_BAR) >= 0;
+    };
+    
+    it('should parse version strings correctly', () => {
+        expect(parseVersion('1.33.9')).toEqual([1, 33, 9]);
+        expect(parseVersion('v1.33.9')).toEqual([1, 33, 9]);
+        expect(parseVersion('1.33.9-beta')).toEqual([1, 33, 9]);
+        expect(parseVersion('1.33')).toEqual([1, 33, 0]);
+        expect(parseVersion('1')).toEqual([1, 0, 0]);
+        expect(parseVersion('')).toEqual([0, 0, 0]);
+        expect(parseVersion(null)).toEqual([0, 0, 0]);
+    });
+    
+    it('should compare versions correctly', () => {
+        expect(compareVersions('1.33.9', '1.33.9')).toBe(0);
+        expect(compareVersions('1.33.10', '1.33.9')).toBe(1);
+        expect(compareVersions('1.34.0', '1.33.9')).toBe(1);
+        expect(compareVersions('2.0.0', '1.33.9')).toBe(1);
+        expect(compareVersions('1.33.8', '1.33.9')).toBe(-1);
+        expect(compareVersions('1.32.0', '1.33.9')).toBe(-1);
+        expect(compareVersions('0.9.9', '1.33.9')).toBe(-1);
+    });
+    
+    it('should return false for versions below 1.33.9', () => {
+        expect(supportsActionBarButtons('1.33.8')).toBe(false);
+        expect(supportsActionBarButtons('1.32.0')).toBe(false);
+        expect(supportsActionBarButtons('0.9.9')).toBe(false);
+    });
+    
+    it('should return true for versions 1.33.9 and above', () => {
+        expect(supportsActionBarButtons('1.33.9')).toBe(true);
+        expect(supportsActionBarButtons('1.33.10')).toBe(true);
+        expect(supportsActionBarButtons('1.34.0')).toBe(true);
+        expect(supportsActionBarButtons('2.0.0')).toBe(true);
+    });
+    
+    it('should handle edge cases in version parsing', () => {
+        expect(supportsActionBarButtons('v1.33.9')).toBe(true);
+        expect(supportsActionBarButtons('1.33.9-rc.1')).toBe(true);
+        expect(supportsActionBarButtons('1.33.9-beta')).toBe(true);
+    });
+});
--- a/tests/routes/test_recipe_routes.py
+++ b/tests/routes/test_recipe_routes.py
@@ -1,4 +1,5 @@
 """Integration smoke tests for the recipe route stack."""
+
 from __future__ import annotations

 import json
@@ -94,19 +95,25 @@ class StubAnalysisService:
        self._recipe_parser_factory = None
        StubAnalysisService.instances.append(self)

-    async def analyze_uploaded_image(self, *, image_bytes: bytes | None, recipe_scanner) -> SimpleNamespace:  # noqa: D401 - mirrors real signature
+    async def analyze_uploaded_image(
+        self, *, image_bytes: bytes | None, recipe_scanner
+    ) -> SimpleNamespace:  # noqa: D401 - mirrors real signature
        if self.raise_for_uploaded:
            raise self.raise_for_uploaded
        self.upload_calls.append(image_bytes or b"")
        return self.result

-    async def analyze_remote_image(self, *, url: Optional[str], recipe_scanner, civitai_client) -> SimpleNamespace:  # noqa: D401
+    async def analyze_remote_image(
+        self, *, url: Optional[str], recipe_scanner, civitai_client
+    ) -> SimpleNamespace:  # noqa: D401
        if self.raise_for_remote:
            raise self.raise_for_remote
        self.remote_calls.append(url)
        return self.result

-    async def analyze_local_image(self, *, file_path: Optional[str], recipe_scanner) -> SimpleNamespace:  # noqa: D401
+    async def analyze_local_image(
+        self, *, file_path: Optional[str], recipe_scanner
+    ) -> SimpleNamespace:  # noqa: D401
        if self.raise_for_local:
            raise self.raise_for_local
        self.local_calls.append(file_path)
@@ -125,11 +132,23 @@ class StubPersistenceService:
        self.save_calls: List[Dict[str, Any]] = []
        self.delete_calls: List[str] = []
        self.move_calls: List[Dict[str, str]] = []
-        self.save_result = SimpleNamespace(payload={"success": True, "recipe_id": "stub-id"}, status=200)
+        self.save_result = SimpleNamespace(
+            payload={"success": True, "recipe_id": "stub-id"}, status=200
+        )
        self.delete_result = SimpleNamespace(payload={"success": True}, status=200)
        StubPersistenceService.instances.append(self)

-    async def save_recipe(self, *, recipe_scanner, image_bytes, image_base64, name, tags, metadata, extension=None) -> SimpleNamespace:  # noqa: D401
+    async def save_recipe(
+        self,
+        *,
+        recipe_scanner,
+        image_bytes,
+        image_base64,
+        name,
+        tags,
+        metadata,
+        extension=None,
+    ) -> SimpleNamespace:  # noqa: D401
        self.save_calls.append(
            {
                "recipe_scanner": recipe_scanner,
@@ -148,22 +167,42 @@ class StubPersistenceService:
        await recipe_scanner.remove_recipe(recipe_id)
        return self.delete_result

-    async def move_recipe(self, *, recipe_scanner, recipe_id: str, target_path: str) -> SimpleNamespace:  # noqa: D401
+    async def move_recipe(
+        self, *, recipe_scanner, recipe_id: str, target_path: str
+    ) -> SimpleNamespace:  # noqa: D401
        self.move_calls.append({"recipe_id": recipe_id, "target_path": target_path})
        return SimpleNamespace(
-            payload={"success": True, "recipe_id": recipe_id, "new_file_path": target_path}, status=200
+            payload={
+                "success": True,
+                "recipe_id": recipe_id,
+                "new_file_path": target_path,
+            },
+            status=200,
        )

-    async def update_recipe(self, *, recipe_scanner, recipe_id: str, updates: Dict[str, Any]) -> SimpleNamespace:  # pragma: no cover - unused by smoke tests
-        return SimpleNamespace(payload={"success": True, "recipe_id": recipe_id, "updates": updates}, status=200)
+    async def update_recipe(
+        self, *, recipe_scanner, recipe_id: str, updates: Dict[str, Any]
+    ) -> SimpleNamespace:  # pragma: no cover - unused by smoke tests
+        return SimpleNamespace(
+            payload={"success": True, "recipe_id": recipe_id, "updates": updates},
+            status=200,
+        )

-    async def reconnect_lora(self, *, recipe_scanner, recipe_id: str, lora_index: int, target_name: str) -> SimpleNamespace:  # pragma: no cover
+    async def reconnect_lora(
+        self, *, recipe_scanner, recipe_id: str, lora_index: int, target_name: str
+    ) -> SimpleNamespace:  # pragma: no cover
        return SimpleNamespace(payload={"success": True}, status=200)

-    async def bulk_delete(self, *, recipe_scanner, recipe_ids: List[str]) -> SimpleNamespace:  # pragma: no cover
-        return SimpleNamespace(payload={"success": True, "deleted": recipe_ids}, status=200)
+    async def bulk_delete(
+        self, *, recipe_scanner, recipe_ids: List[str]
+    ) -> SimpleNamespace:  # pragma: no cover
+        return SimpleNamespace(
+            payload={"success": True, "deleted": recipe_ids}, status=200
+        )

-    async def save_recipe_from_widget(self, *, recipe_scanner, metadata: Dict[str, Any], image_bytes: bytes) -> SimpleNamespace:  # pragma: no cover
+    async def save_recipe_from_widget(
+        self, *, recipe_scanner, metadata: Dict[str, Any], image_bytes: bytes
+    ) -> SimpleNamespace:  # pragma: no cover
        return SimpleNamespace(payload={"success": True}, status=200)


@@ -176,7 +215,11 @@ class StubSharingService:
        self.share_calls: List[str] = []
        self.download_calls: List[str] = []
        self.share_result = SimpleNamespace(
-            payload={"success": True, "download_url": "/share/stub", "filename": "recipe.png"},
+            payload={
+                "success": True,
+                "download_url": "/share/stub",
+                "filename": "recipe.png",
+            },
            status=200,
        )
        self.download_info = SimpleNamespace(file_path="", download_filename="")
@@ -186,7 +229,9 @@ class StubSharingService:
        self.share_calls.append(recipe_id)
        return self.share_result

-    async def prepare_download(self, *, recipe_scanner, recipe_id: str) -> SimpleNamespace:
+    async def prepare_download(
+        self, *, recipe_scanner, recipe_id: str
+    ) -> SimpleNamespace:
        self.download_calls.append(recipe_id)
        return self.download_info

@@ -214,7 +259,9 @@ class StubCivitaiClient:


@asynccontextmanager
-async def recipe_harness(monkeypatch, tmp_path: Path) -> AsyncIterator[RecipeRouteHarness]:
+async def recipe_harness(
+    monkeypatch, tmp_path: Path
+) -> AsyncIterator[RecipeRouteHarness]:
    """Context manager that yields a fully wired recipe route harness."""

    StubAnalysisService.instances.clear()
@@ -237,8 +284,12 @@ async def recipe_harness(monkeypatch, tmp_path: Path) -> AsyncIterator[RecipeRou

    monkeypatch.setattr(ServiceRegistry, "get_recipe_scanner", fake_get_recipe_scanner)
    monkeypatch.setattr(ServiceRegistry, "get_civitai_client", fake_get_civitai_client)
-    monkeypatch.setattr(base_recipe_routes, "RecipeAnalysisService", StubAnalysisService)
-    monkeypatch.setattr(base_recipe_routes, "RecipePersistenceService", StubPersistenceService)
+    monkeypatch.setattr(
+        base_recipe_routes, "RecipeAnalysisService", StubAnalysisService
+    )
+    monkeypatch.setattr(
+        base_recipe_routes, "RecipePersistenceService", StubPersistenceService
+    )
    monkeypatch.setattr(base_recipe_routes, "RecipeSharingService", StubSharingService)
    monkeypatch.setattr(base_recipe_routes, "get_downloader", fake_get_downloader)
    monkeypatch.setattr(config, "loras_roots", [str(tmp_path)], raising=False)
@@ -294,7 +345,9 @@ async def test_list_recipes_provides_file_urls(monkeypatch, tmp_path: Path) -> N
 async def test_save_and_delete_recipe_round_trip(monkeypatch, tmp_path: Path) -> None:
    async with recipe_harness(monkeypatch, tmp_path) as harness:
        form = FormData()
-        form.add_field("image", b"stub", filename="sample.png", content_type="image/png")
+        form.add_field(
+            "image", b"stub", filename="sample.png", content_type="image/png"
+        )
        form.add_field("name", "Test Recipe")
        form.add_field("tags", json.dumps(["tag-a"]))
        form.add_field("metadata", json.dumps({"loras": []}))
@@ -312,7 +365,9 @@ async def test_save_and_delete_recipe_round_trip(monkeypatch, tmp_path: Path) ->
        assert save_payload["recipe_id"] == "saved-id"
        assert harness.persistence.save_calls[-1]["name"] == "Test Recipe"

-        harness.persistence.delete_result = SimpleNamespace(payload={"success": True}, status=200)
+        harness.persistence.delete_result = SimpleNamespace(
+            payload={"success": True}, status=200
+        )

        delete_response = await harness.client.delete("/api/lm/recipe/saved-id")
        delete_payload = await delete_response.json()
@@ -326,14 +381,20 @@ async def test_move_recipe_invokes_persistence(monkeypatch, tmp_path: Path) -> N
    async with recipe_harness(monkeypatch, tmp_path) as harness:
        response = await harness.client.post(
            "/api/lm/recipe/move",
-            json={"recipe_id": "move-me", "target_path": str(tmp_path / "recipes" / "subdir")},
+            json={
+                "recipe_id": "move-me",
+                "target_path": str(tmp_path / "recipes" / "subdir"),
+            },
        )

        payload = await response.json()
        assert response.status == 200
        assert payload["recipe_id"] == "move-me"
        assert harness.persistence.move_calls == [
-            {"recipe_id": "move-me", "target_path": str(tmp_path / "recipes" / "subdir")}
+            {
+                "recipe_id": "move-me",
+                "target_path": str(tmp_path / "recipes" / "subdir"),
+            }
        ]


@@ -348,7 +409,10 @@ async def test_import_remote_recipe(monkeypatch, tmp_path: Path) -> None:
    async def fake_get_default_metadata_provider():
        return Provider()

-    monkeypatch.setattr("py.recipes.enrichment.get_default_metadata_provider", fake_get_default_metadata_provider)
+    monkeypatch.setattr(
+        "py.recipes.enrichment.get_default_metadata_provider",
+        fake_get_default_metadata_provider,
+    )

    async with recipe_harness(monkeypatch, tmp_path) as harness:
        resources = [
@@ -397,7 +461,9 @@ async def test_import_remote_recipe(monkeypatch, tmp_path: Path) -> None:
        assert harness.downloader.urls == ["https://example.com/images/1"]


-async def test_import_remote_recipe_falls_back_to_request_base_model(monkeypatch, tmp_path: Path) -> None:
+async def test_import_remote_recipe_falls_back_to_request_base_model(
+    monkeypatch, tmp_path: Path
+) -> None:
    provider_calls: list[str | int] = []

    class Provider:
@@ -408,7 +474,10 @@ async def test_import_remote_recipe_falls_back_to_request_base_model(monkeypatch
    async def fake_get_default_metadata_provider():
        return Provider()

-    monkeypatch.setattr("py.recipes.enrichment.get_default_metadata_provider", fake_get_default_metadata_provider)
+    monkeypatch.setattr(
+        "py.recipes.enrichment.get_default_metadata_provider",
+        fake_get_default_metadata_provider,
+    )

    async with recipe_harness(monkeypatch, tmp_path) as harness:
        resources = [
@@ -444,13 +513,16 @@ async def test_import_remote_video_recipe(monkeypatch, tmp_path: Path) -> None:
    async def fake_get_default_metadata_provider():
        return SimpleNamespace(get_model_version_info=lambda id: ({}, None))

-    monkeypatch.setattr("py.recipes.enrichment.get_default_metadata_provider", fake_get_default_metadata_provider)
+    monkeypatch.setattr(
+        "py.recipes.enrichment.get_default_metadata_provider",
+        fake_get_default_metadata_provider,
+    )

    async with recipe_harness(monkeypatch, tmp_path) as harness:
        harness.civitai.image_info["12345"] = {
            "id": 12345,
            "url": "https://image.civitai.com/x/y/original=true/video.mp4",
-            "type": "video"
+            "type": "video",
        }

        response = await harness.client.get(
@@ -469,7 +541,7 @@ async def test_import_remote_video_recipe(monkeypatch, tmp_path: Path) -> None:

        # Verify downloader was called with rewritten URL
        assert "transcode=true" in harness.downloader.urls[0]
-        
+
        # Verify persistence was called with correct extension
        call = harness.persistence.save_calls[-1]
        assert call["extension"] == ".mp4"
@@ -477,7 +549,9 @@ async def test_import_remote_video_recipe(monkeypatch, tmp_path: Path) -> None:

 async def test_analyze_uploaded_image_error_path(monkeypatch, tmp_path: Path) -> None:
    async with recipe_harness(monkeypatch, tmp_path) as harness:
-        harness.analysis.raise_for_uploaded = RecipeValidationError("No image data provided")
+        harness.analysis.raise_for_uploaded = RecipeValidationError(
+            "No image data provided"
+        )

        form = FormData()
        form.add_field("image", b"", filename="empty.png", content_type="image/png")
@@ -504,7 +578,11 @@ async def test_share_and_download_recipe(monkeypatch, tmp_path: Path) -> None:
        }

        harness.sharing.share_result = SimpleNamespace(
-            payload={"success": True, "download_url": "/api/share", "filename": "share.png"},
+            payload={
+                "success": True,
+                "download_url": "/api/share",
+                "filename": "share.png",
+            },
            status=200,
        )
        harness.sharing.download_info = SimpleNamespace(
@@ -519,15 +597,24 @@ async def test_share_and_download_recipe(monkeypatch, tmp_path: Path) -> None:
        assert share_payload["filename"] == "share.png"
        assert harness.sharing.share_calls == [recipe_id]

-        download_response = await harness.client.get(f"/api/lm/recipe/{recipe_id}/share/download")
+        download_response = await harness.client.get(
+            f"/api/lm/recipe/{recipe_id}/share/download"
+        )
        body = await download_response.read()

        assert download_response.status == 200
-        assert download_response.headers["Content-Disposition"] == 'attachment; filename="share.png"'
+        assert (
+            download_response.headers["Content-Disposition"]
+            == 'attachment; filename="share.png"'
+        )
        assert body == b"stub"

        download_path.unlink(missing_ok=True)
-async def test_import_remote_recipe_merges_metadata(monkeypatch, tmp_path: Path) -> None:
+
+
+async def test_import_remote_recipe_merges_metadata(
+    monkeypatch, tmp_path: Path
+) -> None:
    # 1. Mock Metadata Provider
    class Provider:
        async def get_model_version_info(self, model_version_id):
@@ -536,22 +623,25 @@ async def test_import_remote_recipe_merges_metadata(monkeypatch, tmp_path: Path)
    async def fake_get_default_metadata_provider():
        return Provider()

-    monkeypatch.setattr("py.recipes.enrichment.get_default_metadata_provider", fake_get_default_metadata_provider)
+    monkeypatch.setattr(
+        "py.recipes.enrichment.get_default_metadata_provider",
+        fake_get_default_metadata_provider,
+    )

    # 2. Mock ExifUtils to return some embedded metadata
    class MockExifUtils:
        @staticmethod
        def extract_image_metadata(path):
-            return "Recipe metadata: " + json.dumps({
-                "gen_params": {"prompt": "from embedded", "seed": 123}
-            })
+            return "Recipe metadata: " + json.dumps(
+                {"gen_params": {"prompt": "from embedded", "seed": 123}}
+            )

    monkeypatch.setattr(recipe_handlers, "ExifUtils", MockExifUtils)

    # 3. Mock Parser Factory for StubAnalysisService
    class MockParser:
        async def parse_metadata(self, raw, recipe_scanner=None):
-            return json.loads(raw[len("Recipe metadata: "):])
+            return json.loads(raw[len("Recipe metadata: ") :])

    class MockFactory:
        def create_parser(self, raw):
@@ -562,12 +652,12 @@ async def test_import_remote_recipe_merges_metadata(monkeypatch, tmp_path: Path)
    # 4. Setup Harness and run test
    async with recipe_harness(monkeypatch, tmp_path) as harness:
        harness.analysis._recipe_parser_factory = MockFactory()
-        
+
        # Civitai meta via image_info
        harness.civitai.image_info["1"] = {
            "id": 1,
            "url": "https://example.com/images/1.jpg",
-            "meta": {"prompt": "from civitai", "cfg": 7.0}
+            "meta": {"prompt": "from civitai", "cfg": 7.0},
        }

        resources = []
@@ -583,11 +673,11 @@ async def test_import_remote_recipe_merges_metadata(monkeypatch, tmp_path: Path)

        payload = await response.json()
        assert response.status == 200
-        
+
        call = harness.persistence.save_calls[-1]
        metadata = call["metadata"]
        gen_params = metadata["gen_params"]
-        
+
        assert gen_params["seed"] == 123


@@ -619,3 +709,142 @@ async def test_get_recipe_syntax(monkeypatch, tmp_path: Path) -> None:
        response_404 = await harness.client.get("/api/lm/recipe/non-existent/syntax")
        assert response_404.status == 404

+
+async def test_batch_import_start_success(monkeypatch, tmp_path: Path) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        response = await harness.client.post(
+            "/api/lm/recipes/batch-import/start",
+            json={
+                "items": [
+                    {"source": "https://example.com/image1.png"},
+                    {"source": "https://example.com/image2.png"},
+                ],
+                "tags": ["batch", "import"],
+                "skip_no_metadata": True,
+            },
+        )
+        payload = await response.json()
+        assert response.status == 200
+        assert payload["success"] is True
+        assert "operation_id" in payload
+
+
+async def test_batch_import_start_empty_items(monkeypatch, tmp_path: Path) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        response = await harness.client.post(
+            "/api/lm/recipes/batch-import/start",
+            json={"items": [], "tags": []},
+        )
+        payload = await response.json()
+        assert response.status == 400
+        assert payload["success"] is False
+        assert "No items provided" in payload["error"]
+
+
+async def test_batch_import_start_missing_source(monkeypatch, tmp_path: Path) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        response = await harness.client.post(
+            "/api/lm/recipes/batch-import/start",
+            json={"items": [{"source": ""}]},
+        )
+        payload = await response.json()
+        assert response.status == 400
+        assert payload["success"] is False
+        assert "source" in payload["error"].lower()
+
+
+async def test_batch_import_start_already_running(monkeypatch, tmp_path: Path) -> None:
+    import asyncio
+
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        original_analyze = harness.analysis.analyze_remote_image
+
+        async def slow_analyze(*, url, recipe_scanner, civitai_client):
+            await asyncio.sleep(0.5)
+            return await original_analyze(
+                url=url, recipe_scanner=recipe_scanner, civitai_client=civitai_client
+            )
+
+        harness.analysis.analyze_remote_image = slow_analyze
+
+        items = [{"source": f"https://example.com/image{i}.png"} for i in range(10)]
+
+        response1 = await harness.client.post(
+            "/api/lm/recipes/batch-import/start",
+            json={"items": items},
+        )
+        assert response1.status == 200
+
+        payload1 = await response1.json()
+        assert payload1["success"] is True
+
+        await asyncio.sleep(0.1)
+
+        response2 = await harness.client.post(
+            "/api/lm/recipes/batch-import/start",
+            json={"items": [{"source": "https://example.com/other.png"}]},
+        )
+        payload2 = await response2.json()
+        assert response2.status == 409
+        assert "already in progress" in payload2["error"].lower()
+
+
+async def test_batch_import_get_progress_not_found(monkeypatch, tmp_path: Path) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        response = await harness.client.get(
+            "/api/lm/recipes/batch-import/progress",
+            params={"operation_id": "nonexistent-id"},
+        )
+        payload = await response.json()
+        assert response.status == 404
+        assert payload["success"] is False
+
+
+async def test_batch_import_get_progress_missing_id(
+    monkeypatch, tmp_path: Path
+) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        response = await harness.client.get("/api/lm/recipes/batch-import/progress")
+        payload = await response.json()
+        assert response.status == 400
+        assert payload["success"] is False
+
+
+async def test_batch_import_cancel_success(monkeypatch, tmp_path: Path) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        start_response = await harness.client.post(
+            "/api/lm/recipes/batch-import/start",
+            json={"items": [{"source": "https://example.com/image.png"}]},
+        )
+        start_payload = await start_response.json()
+        operation_id = start_payload["operation_id"]
+
+        cancel_response = await harness.client.post(
+            "/api/lm/recipes/batch-import/cancel",
+            json={"operation_id": operation_id},
+        )
+        cancel_payload = await cancel_response.json()
+        assert cancel_response.status == 200
+        assert cancel_payload["success"] is True
+
+
+async def test_batch_import_cancel_not_found(monkeypatch, tmp_path: Path) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        response = await harness.client.post(
+            "/api/lm/recipes/batch-import/cancel",
+            json={"operation_id": "nonexistent-id"},
+        )
+        payload = await response.json()
+        assert response.status == 404
+        assert payload["success"] is False
+
+
+async def test_batch_import_cancel_missing_id(monkeypatch, tmp_path: Path) -> None:
+    async with recipe_harness(monkeypatch, tmp_path) as harness:
+        response = await harness.client.post(
+            "/api/lm/recipes/batch-import/cancel",
+            json={},
+        )
+        payload = await response.json()
+        assert response.status == 400
+        assert payload["success"] is False
--- a/tests/services/test_batch_import_service.py
+++ b/tests/services/test_batch_import_service.py
@@ -0,0 +1,597 @@
+"""Unit tests for BatchImportService."""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+import os
+import tempfile
+from dataclasses import dataclass
+from pathlib import Path
+from types import SimpleNamespace
+from typing import Any, Dict, List, Optional
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+
+from py.services.batch_import_service import (
+    AdaptiveConcurrencyController,
+    BatchImportItem,
+    BatchImportProgress,
+    BatchImportService,
+    ImportItemType,
+    ImportStatus,
+)
+
+
+class MockWebSocketManager:
+    def __init__(self):
+        self.broadcasts: List[Dict[str, Any]] = []
+
+    async def broadcast(self, data: Dict[str, Any]):
+        self.broadcasts.append(data)
+
+
+@dataclass
+class MockAnalysisResult:
+    payload: Dict[str, Any]
+    status: int = 200
+
+
+class MockAnalysisService:
+    def __init__(self, results: Optional[Dict[str, MockAnalysisResult]] = None):
+        self.results = results or {}
+        self.call_count = 0
+        self.last_url = None
+        self.last_path = None
+
+    async def analyze_remote_image(self, *, url: str, recipe_scanner, civitai_client):
+        self.call_count += 1
+        self.last_url = url
+        if url in self.results:
+            return self.results[url]
+        return MockAnalysisResult({"error": "No metadata found", "loras": []})
+
+    async def analyze_local_image(self, *, file_path: str, recipe_scanner):
+        self.call_count += 1
+        self.last_path = file_path
+        if file_path in self.results:
+            return self.results[file_path]
+        return MockAnalysisResult({"error": "No metadata found", "loras": []})
+
+
+@dataclass
+class MockSaveResult:
+    payload: Dict[str, Any]
+    status: int = 200
+
+
+class MockPersistenceService:
+    def __init__(self, should_succeed: bool = True):
+        self.should_succeed = should_succeed
+        self.saved_recipes: List[Dict[str, Any]] = []
+        self.call_count = 0
+
+    async def save_recipe(
+        self,
+        *,
+        recipe_scanner,
+        image_bytes: Optional[bytes] = None,
+        image_base64: Optional[str] = None,
+        name: str,
+        tags: List[str],
+        metadata: Dict[str, Any],
+        extension: Optional[str] = None,
+    ):
+        self.call_count += 1
+        self.saved_recipes.append(
+            {
+                "name": name,
+                "tags": tags,
+                "metadata": metadata,
+            }
+        )
+        if self.should_succeed:
+            return MockSaveResult({"success": True, "id": f"recipe_{self.call_count}"})
+        return MockSaveResult({"success": False, "error": "Save failed"}, status=400)
+
+
+class TestAdaptiveConcurrencyController:
+    def test_initial_values(self):
+        controller = AdaptiveConcurrencyController()
+        assert controller.current_concurrency == 3
+        assert controller.min_concurrency == 1
+        assert controller.max_concurrency == 5
+
+    def test_custom_initial_values(self):
+        controller = AdaptiveConcurrencyController(
+            min_concurrency=2,
+            max_concurrency=10,
+            initial_concurrency=5,
+        )
+        assert controller.current_concurrency == 5
+        assert controller.min_concurrency == 2
+        assert controller.max_concurrency == 10
+
+    def test_increase_concurrency_on_success(self):
+        controller = AdaptiveConcurrencyController(initial_concurrency=3)
+        controller.record_result(duration=0.5, success=True)
+        assert controller.current_concurrency == 4
+
+    def test_do_not_exceed_max(self):
+        controller = AdaptiveConcurrencyController(
+            max_concurrency=5,
+            initial_concurrency=5,
+        )
+        controller.record_result(duration=0.5, success=True)
+        assert controller.current_concurrency == 5
+
+    def test_decrease_concurrency_on_failure(self):
+        controller = AdaptiveConcurrencyController(initial_concurrency=3)
+        controller.record_result(duration=1.0, success=False)
+        assert controller.current_concurrency == 2
+
+    def test_do_not_go_below_min(self):
+        controller = AdaptiveConcurrencyController(
+            min_concurrency=1,
+            initial_concurrency=1,
+        )
+        controller.record_result(duration=1.0, success=False)
+        assert controller.current_concurrency == 1
+
+    def test_slow_task_decreases_concurrency(self):
+        controller = AdaptiveConcurrencyController(initial_concurrency=3)
+        controller.record_result(duration=11.0, success=True)
+        assert controller.current_concurrency == 2
+
+    def test_fast_task_increases_concurrency(self):
+        controller = AdaptiveConcurrencyController(initial_concurrency=3)
+        controller.record_result(duration=0.5, success=True)
+        assert controller.current_concurrency == 4
+
+    def test_moderate_task_no_change(self):
+        controller = AdaptiveConcurrencyController(initial_concurrency=3)
+        controller.record_result(duration=5.0, success=True)
+        assert controller.current_concurrency == 3
+
+
+class TestBatchImportProgress:
+    def test_to_dict(self):
+        progress = BatchImportProgress(
+            operation_id="test-123",
+            total=10,
+            completed=5,
+            success=3,
+            failed=2,
+            skipped=0,
+            current_item="image.png",
+            status="running",
+        )
+        result = progress.to_dict()
+        assert result["operation_id"] == "test-123"
+        assert result["total"] == 10
+        assert result["completed"] == 5
+        assert result["success"] == 3
+        assert result["failed"] == 2
+        assert result["progress_percent"] == 50.0
+
+    def test_progress_percent_zero_total(self):
+        progress = BatchImportProgress(
+            operation_id="test-123",
+            total=0,
+        )
+        assert progress.to_dict()["progress_percent"] == 0
+
+
+class TestBatchImportItem:
+    def test_defaults(self):
+        item = BatchImportItem(
+            id="item-1",
+            source="https://example.com/image.png",
+            item_type=ImportItemType.URL,
+        )
+        assert item.status == ImportStatus.PENDING
+        assert item.error_message is None
+        assert item.recipe_name is None
+
+
+class TestBatchImportService:
+    @pytest.fixture
+    def mock_services(self):
+        ws_manager = MockWebSocketManager()
+        analysis_service = MockAnalysisService()
+        persistence_service = MockPersistenceService()
+        logger = logging.getLogger("test")
+        return ws_manager, analysis_service, persistence_service, logger
+
+    @pytest.fixture
+    def service(self, mock_services):
+        ws_manager, analysis_service, persistence_service, logger = mock_services
+        return BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+
+    def test_is_import_running_no_operations(self, service):
+        assert not service.is_import_running()
+
+    @pytest.mark.asyncio
+    async def test_start_batch_import_creates_operation(self, service):
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[{"source": "https://example.com/image.png"}],
+        )
+
+        assert operation_id is not None
+        assert service.is_import_running(operation_id)
+
+    @pytest.mark.asyncio
+    async def test_get_progress(self, service):
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[
+                {"source": "https://example.com/1.png"},
+                {"source": "https://example.com/2.png"},
+            ],
+        )
+
+        progress = service.get_progress(operation_id)
+        assert progress is not None
+        assert progress.total == 2
+        assert progress.status in ("pending", "running")
+
+    @pytest.mark.asyncio
+    async def test_cancel_import(self, service):
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[{"source": "https://example.com/image.png"}],
+        )
+
+        assert service.cancel_import(operation_id) is True
+        assert service.cancel_import("nonexistent") is False
+
+    @pytest.mark.asyncio
+    async def test_discover_images_non_recursive(self, service, tmp_path):
+        for i in range(3):
+            (tmp_path / f"image{i}.png").write_bytes(b"fake-image")
+
+        (tmp_path / "subdir").mkdir()
+        (tmp_path / "subdir" / "hidden.png").write_bytes(b"fake-image")
+
+        images = await service._discover_images(str(tmp_path), recursive=False)
+        assert len(images) == 3
+
+    @pytest.mark.asyncio
+    async def test_discover_images_recursive(self, service, tmp_path):
+        for i in range(2):
+            (tmp_path / f"image{i}.png").write_bytes(b"fake-image")
+
+        subdir = tmp_path / "subdir"
+        subdir.mkdir()
+        for i in range(2):
+            (subdir / f"nested{i}.jpg").write_bytes(b"fake-image")
+
+        images = await service._discover_images(str(tmp_path), recursive=True)
+        assert len(images) == 4
+
+    @pytest.mark.asyncio
+    async def test_discover_images_filters_by_extension(self, service, tmp_path):
+        (tmp_path / "image.png").write_bytes(b"fake-image")
+        (tmp_path / "image.jpg").write_bytes(b"fake-image")
+        (tmp_path / "image.webp").write_bytes(b"fake-image")
+        (tmp_path / "document.pdf").write_bytes(b"fake-doc")
+        (tmp_path / "script.py").write_bytes(b"print('hello')")
+
+        images = await service._discover_images(str(tmp_path), recursive=False)
+        assert len(images) == 3
+
+    @pytest.mark.asyncio
+    async def test_discover_images_invalid_directory(self, service):
+        from py.services.recipes.errors import RecipeValidationError
+
+        with pytest.raises(RecipeValidationError):
+            await service._discover_images("/nonexistent/path", recursive=False)
+
+    def test_is_supported_image(self, service):
+        assert service._is_supported_image("test.png") is True
+        assert service._is_supported_image("test.jpg") is True
+        assert service._is_supported_image("test.jpeg") is True
+        assert service._is_supported_image("test.webp") is True
+        assert service._is_supported_image("test.gif") is True
+        assert service._is_supported_image("test.bmp") is True
+        assert service._is_supported_image("test.pdf") is False
+        assert service._is_supported_image("test.txt") is False
+
+    @pytest.mark.asyncio
+    async def test_batch_import_processes_items(self, mock_services, tmp_path):
+        ws_manager, _, persistence_service, logger = mock_services
+
+        analysis_service = MockAnalysisService(
+            {
+                "https://example.com/valid.png": MockAnalysisResult(
+                    {
+                        "loras": [{"name": "test-lora", "weight": 1.0}],
+                        "base_model": "SD1.5",
+                        "gen_params": {"steps": 20},
+                    }
+                ),
+            }
+        )
+
+        service = BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+
+        recipe_scanner_getter = lambda: SimpleNamespace(
+            find_recipes_by_fingerprint=lambda x: [],
+            add_recipe=lambda x: None,
+        )
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[
+                {"source": "https://example.com/valid.png"},
+                {"source": "https://example.com/no-meta.png"},
+            ],
+            skip_no_metadata=True,
+        )
+
+        await asyncio.sleep(0.5)
+
+        progress = service.get_progress(operation_id)
+        assert progress is not None or persistence_service.call_count == 1
+
+    @pytest.mark.asyncio
+    async def test_start_directory_import(self, service, tmp_path):
+        for i in range(5):
+            (tmp_path / f"image{i}.png").write_bytes(b"fake-image")
+
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_directory_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            directory=str(tmp_path),
+            recursive=False,
+        )
+
+        progress = service.get_progress(operation_id)
+        assert progress is not None
+        assert progress.total == 5
+
+    @pytest.mark.asyncio
+    async def test_websocket_broadcasts_progress(self, mock_services):
+        ws_manager, analysis_service, persistence_service, logger = mock_services
+
+        service = BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[{"source": "https://example.com/test.png"}],
+        )
+
+        await asyncio.sleep(0.3)
+
+        assert len(ws_manager.broadcasts) > 0
+        assert any(
+            b.get("type") == "batch_import_progress" for b in ws_manager.broadcasts
+        )
+
+    @pytest.mark.asyncio
+    async def test_cancellation_stops_processing(self, mock_services):
+        ws_manager, analysis_service, persistence_service, logger = mock_services
+
+        service = BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        items = [{"source": f"https://example.com/{i}.png"} for i in range(10)]
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=items,
+        )
+
+        service.cancel_import(operation_id)
+        await asyncio.sleep(0.3)
+
+        progress = service.get_progress(operation_id)
+        if progress:
+            assert progress.status == "cancelled"
+
+
+class TestBatchImportServiceEdgeCases:
+    @pytest.fixture
+    def service(self):
+        ws_manager = MockWebSocketManager()
+        analysis_service = MockAnalysisService()
+        persistence_service = MockPersistenceService()
+        logger = logging.getLogger("test")
+
+        return BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+
+    @pytest.mark.asyncio
+    async def test_empty_items_list(self, service):
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[],
+        )
+
+        progress = service.get_progress(operation_id)
+        assert progress is not None
+        assert progress.total == 0
+
+    @pytest.mark.asyncio
+    async def test_mixed_url_and_path_items(self, service, tmp_path):
+        (tmp_path / "local.png").write_bytes(b"fake-image")
+
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[
+                {"source": "https://example.com/remote.png", "type": "url"},
+                {"source": str(tmp_path / "local.png"), "type": "local_path"},
+            ],
+        )
+
+        progress = service.get_progress(operation_id)
+        assert progress is not None
+        assert progress.total == 2
+        assert progress.items[0].item_type == ImportItemType.URL
+        assert progress.items[1].item_type == ImportItemType.LOCAL_PATH
+
+    @pytest.mark.asyncio
+    async def test_tags_are_passed_to_persistence(self, tmp_path):
+        ws_manager = MockWebSocketManager()
+        analysis_service = MockAnalysisService(
+            {
+                str(tmp_path / "test.png"): MockAnalysisResult(
+                    {
+                        "loras": [{"name": "test-lora"}],
+                    }
+                ),
+            }
+        )
+        persistence_service = MockPersistenceService()
+        logger = logging.getLogger("test")
+
+        (tmp_path / "test.png").write_bytes(b"fake-image")
+
+        service = BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+
+        recipe_scanner_getter = lambda: SimpleNamespace(
+            find_recipes_by_fingerprint=lambda x: [],
+        )
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[{"source": str(tmp_path / "test.png")}],
+            tags=["batch-import", "test"],
+        )
+
+        await asyncio.sleep(0.3)
+
+        if persistence_service.saved_recipes:
+            assert "batch-import" in persistence_service.saved_recipes[0]["tags"]
+            assert "test" in persistence_service.saved_recipes[0]["tags"]
+
+    @pytest.mark.asyncio
+    async def test_skip_duplicates_parameter(self, service):
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[{"source": "https://example.com/test.png"}],
+            skip_duplicates=True,
+        )
+
+        progress = service.get_progress(operation_id)
+        assert progress is not None
+        assert progress.skip_duplicates is True
+
+    @pytest.mark.asyncio
+    async def test_skip_duplicates_false_by_default(self, service):
+        recipe_scanner_getter = lambda: SimpleNamespace()
+        civitai_client_getter = lambda: SimpleNamespace()
+
+        operation_id = await service.start_batch_import(
+            recipe_scanner_getter=recipe_scanner_getter,
+            civitai_client_getter=civitai_client_getter,
+            items=[{"source": "https://example.com/test.png"}],
+        )
+
+        progress = service.get_progress(operation_id)
+        assert progress is not None
+        assert progress.skip_duplicates is False
+
+
+class TestInputValidation:
+    @pytest.fixture
+    def service(self):
+        ws_manager = MockWebSocketManager()
+        analysis_service = MockAnalysisService()
+        persistence_service = MockPersistenceService()
+        logger = logging.getLogger("test")
+
+        return BatchImportService(
+            analysis_service=analysis_service,
+            persistence_service=persistence_service,
+            ws_manager=ws_manager,
+            logger=logger,
+        )
+
+    def test_validate_valid_url(self, service):
+        assert service._validate_url("https://example.com/image.png") is True
+        assert service._validate_url("http://example.com/image.png") is True
+        assert service._validate_url("https://civitai.com/images/123") is True
+
+    def test_validate_invalid_url(self, service):
+        assert service._validate_url("not-a-url") is False
+        assert service._validate_url("ftp://example.com/file") is False
+        assert service._validate_url("") is False
+
+    def test_validate_valid_local_path(self, service, tmp_path):
+        valid_path = str(tmp_path / "image.png")
+        assert service._validate_local_path(valid_path) is True
+
+    def test_validate_invalid_local_path(self, service):
+        assert service._validate_local_path("../etc/passwd") is False
+        assert service._validate_local_path("relative/path.png") is False
+        assert service._validate_local_path("") is False
--- a/tests/services/test_recipe_scanner.py
+++ b/tests/services/test_recipe_scanner.py
@@ -60,11 +60,13 @@ class StubLoraScanner:
                "preview_url": info.get("preview_url", ""),
                "civitai": info.get("civitai", {}),
            }
-        self._cache.raw_data.append({
-            "sha256": info.get("sha256", ""),
-            "path": info.get("file_path", ""),
-            "civitai": info.get("civitai", {}),
-        })
+        self._cache.raw_data.append(
+            {
+                "sha256": info.get("sha256", ""),
+                "path": info.get("file_path", ""),
+                "civitai": info.get("civitai", {}),
+            }
+        )


@pytest.fixture
@@ -107,7 +109,8 @@ async def test_add_recipe_during_concurrent_reads(recipe_scanner):
            await asyncio.sleep(0)

    await asyncio.gather(reader_task(), reader_task(), scanner.add_recipe(new_recipe))
-    await asyncio.sleep(0)
+    # Wait a bit longer for the thread-pool resort to complete
+    await asyncio.sleep(0.1)
    cache = await scanner.get_cached_data()

    assert {item["id"] for item in cache.raw_data} == {"one", "two"}
@@ -119,14 +122,16 @@ async def test_remove_recipe_during_reads(recipe_scanner):

    recipe_ids = ["alpha", "beta", "gamma"]
    for index, recipe_id in enumerate(recipe_ids):
-        await scanner.add_recipe({
-            "id": recipe_id,
-            "file_path": f"path/{recipe_id}.png",
-            "title": recipe_id,
-            "modified": float(index),
-            "created_date": float(index),
-            "loras": [],
-        })
+        await scanner.add_recipe(
+            {
+                "id": recipe_id,
+                "file_path": f"path/{recipe_id}.png",
+                "title": recipe_id,
+                "modified": float(index),
+                "created_date": float(index),
+                "loras": [],
+            }
+        )

    async def reader_task():
        for _ in range(5):
@@ -155,7 +160,13 @@ async def test_update_lora_entry_updates_cache_and_file(tmp_path: Path, recipe_s
        "modified": 0.0,
        "created_date": 0.0,
        "loras": [
-            {"file_name": "old", "strength": 1.0, "hash": "", "isDeleted": True, "exclude": True},
+            {
+                "file_name": "old",
+                "strength": 1.0,
+                "hash": "",
+                "isDeleted": True,
+                "exclude": True,
+            },
        ],
    }
    recipe_path.write_text(json.dumps(recipe_data))
@@ -380,7 +391,9 @@ async def test_initialize_waits_for_lora_scanner(monkeypatch):


@pytest.mark.asyncio
-async def test_invalid_model_version_marked_deleted_and_not_retried(monkeypatch, recipe_scanner):
+async def test_invalid_model_version_marked_deleted_and_not_retried(
+    monkeypatch, recipe_scanner
+):
    scanner, _ = recipe_scanner
    recipes_dir = Path(config.loras_roots[0]) / "recipes"
    recipes_dir.mkdir(parents=True, exist_ok=True)
@@ -417,7 +430,9 @@ async def test_invalid_model_version_marked_deleted_and_not_retried(monkeypatch,


@pytest.mark.asyncio
-async def test_load_recipe_persists_deleted_flag_on_invalid_version(monkeypatch, recipe_scanner, tmp_path):
+async def test_load_recipe_persists_deleted_flag_on_invalid_version(
+    monkeypatch, recipe_scanner, tmp_path
+):
    scanner, _ = recipe_scanner
    recipes_dir = Path(config.loras_roots[0]) / "recipes"
    recipes_dir.mkdir(parents=True, exist_ok=True)
@@ -448,7 +463,9 @@ async def test_load_recipe_persists_deleted_flag_on_invalid_version(monkeypatch,


@pytest.mark.asyncio
-async def test_update_lora_filename_by_hash_updates_affected_recipes(tmp_path: Path, recipe_scanner):
+async def test_update_lora_filename_by_hash_updates_affected_recipes(
+    tmp_path: Path, recipe_scanner
+):
    scanner, _ = recipe_scanner
    recipes_dir = Path(config.loras_roots[0]) / "recipes"
    recipes_dir.mkdir(parents=True, exist_ok=True)
@@ -464,7 +481,7 @@ async def test_update_lora_filename_by_hash_updates_affected_recipes(tmp_path: P
        "created_date": 0.0,
        "loras": [
            {"file_name": "old_name", "hash": "hash1"},
-            {"file_name": "other_lora", "hash": "hash2"}
+            {"file_name": "other_lora", "hash": "hash2"},
        ],
    }
    recipe1_path.write_text(json.dumps(recipe1_data))
@@ -479,16 +496,16 @@ async def test_update_lora_filename_by_hash_updates_affected_recipes(tmp_path: P
        "title": "Recipe 2",
        "modified": 0.0,
        "created_date": 0.0,
-        "loras": [
-            {"file_name": "other_lora", "hash": "hash2"}
-        ],
+        "loras": [{"file_name": "other_lora", "hash": "hash2"}],
    }
    recipe2_path.write_text(json.dumps(recipe2_data))
    await scanner.add_recipe(dict(recipe2_data))

    # Update LoRA name for "hash1" (using different case to test normalization)
    new_name = "new_name"
-    file_count, cache_count = await scanner.update_lora_filename_by_hash("HASH1", new_name)
+    file_count, cache_count = await scanner.update_lora_filename_by_hash(
+        "HASH1", new_name
+    )

    assert file_count == 1
    assert cache_count == 1
@@ -510,92 +527,100 @@ async def test_update_lora_filename_by_hash_updates_affected_recipes(tmp_path: P
@pytest.mark.asyncio
 async def test_get_paginated_data_filters_by_favorite(recipe_scanner):
    scanner, _ = recipe_scanner
-    
+
    # Add a normal recipe
-    await scanner.add_recipe({
-        "id": "regular",
-        "file_path": "path/regular.png",
-        "title": "Regular Recipe",
-        "modified": 1.0,
-        "created_date": 1.0,
-        "loras": [],
-    })
-    
+    await scanner.add_recipe(
+        {
+            "id": "regular",
+            "file_path": "path/regular.png",
+            "title": "Regular Recipe",
+            "modified": 1.0,
+            "created_date": 1.0,
+            "loras": [],
+        }
+    )
+
    # Add a favorite recipe
-    await scanner.add_recipe({
-        "id": "favorite",
-        "file_path": "path/favorite.png",
-        "title": "Favorite Recipe",
-        "modified": 2.0,
-        "created_date": 2.0,
-        "loras": [],
-        "favorite": True
-    })
-    
+    await scanner.add_recipe(
+        {
+            "id": "favorite",
+            "file_path": "path/favorite.png",
+            "title": "Favorite Recipe",
+            "modified": 2.0,
+            "created_date": 2.0,
+            "loras": [],
+            "favorite": True,
+        }
+    )
+
    # Wait for cache update (it's async in some places, add_recipe is usually enough but let's be safe)
    await asyncio.sleep(0)
-    
+
    # Test without filter (should return both)
    result_all = await scanner.get_paginated_data(page=1, page_size=10)
    assert len(result_all["items"]) == 2
-    
+
    # Test with favorite filter
-    result_fav = await scanner.get_paginated_data(page=1, page_size=10, filters={"favorite": True})
+    result_fav = await scanner.get_paginated_data(
+        page=1, page_size=10, filters={"favorite": True}
+    )
    assert len(result_fav["items"]) == 1
    assert result_fav["items"][0]["id"] == "favorite"
-    
+
    # Test with favorite filter set to False (should return both or at least not filter if it's the default)
    # Actually our implementation checks if 'favorite' in filters and filters['favorite']
-    result_fav_false = await scanner.get_paginated_data(page=1, page_size=10, filters={"favorite": False})
+    result_fav_false = await scanner.get_paginated_data(
+        page=1, page_size=10, filters={"favorite": False}
+    )
    assert len(result_fav_false["items"]) == 2


@pytest.mark.asyncio
 async def test_get_paginated_data_filters_by_prompt(recipe_scanner):
    scanner, _ = recipe_scanner
-    
+
    # Add a recipe with a specific prompt
-    await scanner.add_recipe({
-        "id": "prompt-recipe",
-        "file_path": "path/prompt.png",
-        "title": "Prompt Recipe",
-        "modified": 1.0,
-        "created_date": 1.0,
-        "loras": [],
-        "gen_params": {
-            "prompt": "a beautiful forest landscape"
+    await scanner.add_recipe(
+        {
+            "id": "prompt-recipe",
+            "file_path": "path/prompt.png",
+            "title": "Prompt Recipe",
+            "modified": 1.0,
+            "created_date": 1.0,
+            "loras": [],
+            "gen_params": {"prompt": "a beautiful forest landscape"},
        }
-    })
-    
+    )
+
    # Add a recipe with a specific negative prompt
-    await scanner.add_recipe({
-        "id": "neg-prompt-recipe",
-        "file_path": "path/neg.png",
-        "title": "Negative Prompt Recipe",
-        "modified": 2.0,
-        "created_date": 2.0,
-        "loras": [],
-        "gen_params": {
-            "negative_prompt": "ugly, blurry mountains"
+    await scanner.add_recipe(
+        {
+            "id": "neg-prompt-recipe",
+            "file_path": "path/neg.png",
+            "title": "Negative Prompt Recipe",
+            "modified": 2.0,
+            "created_date": 2.0,
+            "loras": [],
+            "gen_params": {"negative_prompt": "ugly, blurry mountains"},
        }
-    })
-    
+    )
+
    await asyncio.sleep(0)
-    
+
    # Test search in prompt
    result_prompt = await scanner.get_paginated_data(
        page=1, page_size=10, search="forest", search_options={"prompt": True}
    )
    assert len(result_prompt["items"]) == 1
    assert result_prompt["items"][0]["id"] == "prompt-recipe"
-    
+
    # Test search in negative prompt
    result_neg = await scanner.get_paginated_data(
        page=1, page_size=10, search="mountains", search_options={"prompt": True}
    )
    assert len(result_neg["items"]) == 1
    assert result_neg["items"][0]["id"] == "neg-prompt-recipe"
-    
+
    # Test search disabled (should not find by prompt)
    result_disabled = await scanner.get_paginated_data(
        page=1, page_size=10, search="forest", search_options={"prompt": False}
@@ -606,38 +631,57 @@ async def test_get_paginated_data_filters_by_prompt(recipe_scanner):
@pytest.mark.asyncio
 async def test_get_paginated_data_sorting(recipe_scanner):
    scanner, _ = recipe_scanner
-    
+
    # Add test recipes
    # Recipe A: Name "Alpha", Date 10, LoRAs 2
-    await scanner.add_recipe({
-        "id": "A", "title": "Alpha", "created_date": 10.0,
-        "loras": [{}, {}], "file_path": "a.png"
-    })
+    await scanner.add_recipe(
+        {
+            "id": "A",
+            "title": "Alpha",
+            "created_date": 10.0,
+            "loras": [{}, {}],
+            "file_path": "a.png",
+        }
+    )
    # Recipe B: Name "Beta", Date 20, LoRAs 1
-    await scanner.add_recipe({
-        "id": "B", "title": "Beta", "created_date": 20.0,
-        "loras": [{}], "file_path": "b.png"
-    })
+    await scanner.add_recipe(
+        {
+            "id": "B",
+            "title": "Beta",
+            "created_date": 20.0,
+            "loras": [{}],
+            "file_path": "b.png",
+        }
+    )
    # Recipe C: Name "Gamma", Date 5, LoRAs 3
-    await scanner.add_recipe({
-        "id": "C", "title": "Gamma", "created_date": 5.0,
-        "loras": [{}, {}, {}], "file_path": "c.png"
-    })
-    
+    await scanner.add_recipe(
+        {
+            "id": "C",
+            "title": "Gamma",
+            "created_date": 5.0,
+            "loras": [{}, {}, {}],
+            "file_path": "c.png",
+        }
+    )
+
    await asyncio.sleep(0)
-    
+
    # Test Name DESC: Gamma, Beta, Alpha
    res = await scanner.get_paginated_data(page=1, page_size=10, sort_by="name:desc")
    assert [i["id"] for i in res["items"]] == ["C", "B", "A"]
-    
+
    # Test LoRA Count DESC: Gamma (3), Alpha (2), Beta (1)
-    res = await scanner.get_paginated_data(page=1, page_size=10, sort_by="loras_count:desc")
+    res = await scanner.get_paginated_data(
+        page=1, page_size=10, sort_by="loras_count:desc"
+    )
    assert [i["id"] for i in res["items"]] == ["C", "A", "B"]
-    
+
    # Test LoRA Count ASC: Beta (1), Alpha (2), Gamma (3)
-    res = await scanner.get_paginated_data(page=1, page_size=10, sort_by="loras_count:asc")
+    res = await scanner.get_paginated_data(
+        page=1, page_size=10, sort_by="loras_count:asc"
+    )
    assert [i["id"] for i in res["items"]] == ["B", "A", "C"]
-    
+
    # Test Date ASC: Gamma (5), Alpha (10), Beta (20)
    res = await scanner.get_paginated_data(page=1, page_size=10, sort_by="date:asc")
    assert [i["id"] for i in res["items"]] == ["C", "A", "B"]
--- a/tests/services/test_relative_path_search.py
+++ b/tests/services/test_relative_path_search.py
@@ -62,3 +62,42 @@ async def test_search_relative_paths_excludes_tokens():
    matching = await service.search_relative_paths("flux -detail")

    assert matching == [f"flux{os.sep}keep-me.safetensors"]
+
+
+@pytest.mark.asyncio
+async def test_search_does_not_match_extension():
+    """Searching for 's' or 'safe' should not match .safetensors extension."""
+    scanner = FakeScanner(
+        [
+            {"file_path": "/models/lora1.safetensors"},
+            {"file_path": "/models/lora2.safetensors"},
+            {"file_path": "/models/special-model.safetensors"},  # 's' in filename
+        ],
+        ["/models"],
+    )
+    service = DummyService("stub", scanner, BaseModelMetadata)
+
+    # Searching for 's' should only match 'special-model', not all .safetensors
+    matching = await service.search_relative_paths("s")
+
+    # Should only match 'special-model' because 's' is in the filename
+    assert len(matching) == 1
+    assert "special-model" in matching[0]
+
+
+@pytest.mark.asyncio
+async def test_search_safe_does_not_match_all_files():
+    """Searching for 'safe' should not match .safetensors extension."""
+    scanner = FakeScanner(
+        [
+            {"file_path": "/models/flux.safetensors"},
+            {"file_path": "/models/detail.safetensors"},
+        ],
+        ["/models"],
+    )
+    service = DummyService("stub", scanner, BaseModelMetadata)
+
+    # Searching for 'safe' should return nothing (no file has 'safe' in its name)
+    matching = await service.search_relative_paths("safe")
+
+    assert len(matching) == 0
--- a/tests/test_checkpoint_loaders.py
+++ b/tests/test_checkpoint_loaders.py
@@ -0,0 +1,158 @@
+"""Tests for checkpoint and unet loaders with extra folder paths support"""
+
+import pytest
+import os
+
+
+# Get project root directory (ComfyUI-Lora-Manager folder)
+PROJECT_ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+
+class TestCheckpointLoaderLM:
+    """Test CheckpointLoaderLM node"""
+
+    def test_class_attributes(self):
+        """Test that CheckpointLoaderLM has required class attributes"""
+        # Import in a way that doesn't require ComfyUI
+        import ast
+
+        filepath = os.path.join(PROJECT_ROOT, "py", "nodes", "checkpoint_loader.py")
+
+        with open(filepath, "r") as f:
+            tree = ast.parse(f.read())
+
+        # Find CheckpointLoaderLM class
+        classes = {
+            node.name: node for node in ast.walk(tree) if isinstance(node, ast.ClassDef)
+        }
+        assert "CheckpointLoaderLM" in classes
+
+        cls = classes["CheckpointLoaderLM"]
+
+        # Check for NAME attribute
+        name_attr = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.Assign)
+            and any(t.id == "NAME" for t in n.targets if isinstance(t, ast.Name))
+        ]
+        assert len(name_attr) > 0, "CheckpointLoaderLM should have NAME attribute"
+
+        # Check for CATEGORY attribute
+        cat_attr = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.Assign)
+            and any(t.id == "CATEGORY" for t in n.targets if isinstance(t, ast.Name))
+        ]
+        assert len(cat_attr) > 0, "CheckpointLoaderLM should have CATEGORY attribute"
+
+        # Check for INPUT_TYPES method
+        input_types = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.FunctionDef) and n.name == "INPUT_TYPES"
+        ]
+        assert len(input_types) > 0, "CheckpointLoaderLM should have INPUT_TYPES method"
+
+        # Check for load_checkpoint method
+        load_method = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.FunctionDef) and n.name == "load_checkpoint"
+        ]
+        assert len(load_method) > 0, (
+            "CheckpointLoaderLM should have load_checkpoint method"
+        )
+
+
+class TestUNETLoaderLM:
+    """Test UNETLoaderLM node"""
+
+    def test_class_attributes(self):
+        """Test that UNETLoaderLM has required class attributes"""
+        # Import in a way that doesn't require ComfyUI
+        import ast
+
+        filepath = os.path.join(PROJECT_ROOT, "py", "nodes", "unet_loader.py")
+
+        with open(filepath, "r") as f:
+            tree = ast.parse(f.read())
+
+        # Find UNETLoaderLM class
+        classes = {
+            node.name: node for node in ast.walk(tree) if isinstance(node, ast.ClassDef)
+        }
+        assert "UNETLoaderLM" in classes
+
+        cls = classes["UNETLoaderLM"]
+
+        # Check for NAME attribute
+        name_attr = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.Assign)
+            and any(t.id == "NAME" for t in n.targets if isinstance(t, ast.Name))
+        ]
+        assert len(name_attr) > 0, "UNETLoaderLM should have NAME attribute"
+
+        # Check for CATEGORY attribute
+        cat_attr = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.Assign)
+            and any(t.id == "CATEGORY" for t in n.targets if isinstance(t, ast.Name))
+        ]
+        assert len(cat_attr) > 0, "UNETLoaderLM should have CATEGORY attribute"
+
+        # Check for INPUT_TYPES method
+        input_types = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.FunctionDef) and n.name == "INPUT_TYPES"
+        ]
+        assert len(input_types) > 0, "UNETLoaderLM should have INPUT_TYPES method"
+
+        # Check for load_unet method
+        load_method = [
+            n
+            for n in cls.body
+            if isinstance(n, ast.FunctionDef) and n.name == "load_unet"
+        ]
+        assert len(load_method) > 0, "UNETLoaderLM should have load_unet method"
+
+
+class TestUtils:
+    """Test utility functions"""
+
+    def test_get_checkpoint_info_absolute_exists(self):
+        """Test that get_checkpoint_info_absolute function exists in utils"""
+        import ast
+
+        filepath = os.path.join(PROJECT_ROOT, "py", "utils", "utils.py")
+
+        with open(filepath, "r") as f:
+            tree = ast.parse(f.read())
+
+        functions = [
+            node.name for node in ast.walk(tree) if isinstance(node, ast.FunctionDef)
+        ]
+        assert "get_checkpoint_info_absolute" in functions, (
+            "get_checkpoint_info_absolute should exist"
+        )
+
+    def test_format_model_name_for_comfyui_exists(self):
+        """Test that _format_model_name_for_comfyui function exists in utils"""
+        import ast
+
+        filepath = os.path.join(PROJECT_ROOT, "py", "utils", "utils.py")
+
+        with open(filepath, "r") as f:
+            tree = ast.parse(f.read())
+
+        functions = [
+            node.name for node in ast.walk(tree) if isinstance(node, ast.FunctionDef)
+        ]
+        assert "_format_model_name_for_comfyui" in functions, (
+            "_format_model_name_for_comfyui should exist"
+        )
--- a/tests/test_custom_words_service.py
+++ b/tests/test_custom_words_service.py
@@ -2,7 +2,10 @@

 import pytest

-from py.services.custom_words_service import CustomWordsService, get_custom_words_service
+from py.services.custom_words_service import (
+    CustomWordsService,
+    get_custom_words_service,
+)


 class TestCustomWordsService:
@@ -99,13 +102,19 @@ class MockTagFTSIndex:
        self.called = False
        self._results = [
            {"tag_name": "hatsune_miku", "category": 4, "post_count": 500000},
-            {"tag_name": "hatsune_miku_(vocaloid)", "category": 4, "post_count": 250000},
+            {
+                "tag_name": "hatsune_miku_(vocaloid)",
+                "category": 4,
+                "post_count": 250000,
+            },
        ]

-    def search(self, query, categories=None, limit=20):
+    def search(self, query, categories=None, limit=20, offset=0):
        self.called = True
        if not query:
            return []
        if categories:
-            return [r for r in self._results if r["category"] in categories][:limit]
-        return self._results[:limit]
+            results = [r for r in self._results if r["category"] in categories]
+        else:
+            results = self._results
+        return results[offset : offset + limit]
--- a/tests/test_tag_fts_index.py
+++ b/tests/test_tag_fts_index.py
@@ -31,10 +31,27 @@ def temp_db_path():
@pytest.fixture
 def temp_csv_path():
    """Create a temporary CSV file with test data."""
-    with tempfile.NamedTemporaryFile(mode="w", suffix=".csv", delete=False, encoding="utf-8") as f:
+    with tempfile.NamedTemporaryFile(
+        mode="w", suffix=".csv", delete=False, encoding="utf-8"
+    ) as f:
        # Write test data in the same format as danbooru_e621_merged.csv
        # Format: tag_name,category,post_count,aliases
+        # Include multiple tags starting with "1" to test popularity-based ranking
        f.write('1girl,0,6008644,"1girls,sole_female"\n')
+        f.write('1boy,0,1405457,"1boys,sole_male"\n')
+        f.write('1:1,14,377032,""\n')
+        f.write('16:9,14,152866,""\n')
+        f.write('1other,0,70962,""\n')
+        f.write('16:10,14,14739,""\n')
+        f.write('1990s_(style),0,9369,""\n')
+        f.write('1_eye,0,7179,""\n')
+        f.write('1:2,14,5865,""\n')
+        f.write('1980s_(style),0,5665,""\n')
+        f.write('1koma,0,4384,""\n')
+        f.write('1_horn,0,2122,""\n')
+        f.write('101_dalmatian_street,3,1933,""\n')
+        f.write('1upgobbo,3,1731,""\n')
+        f.write('14:9,14,1038,""\n')
        f.write('highres,5,5256195,"high_res,high_resolution,hires"\n')
        f.write('solo,0,5000954,"alone,female_solo,single"\n')
        f.write('hatsune_miku,4,500000,"miku"\n')
@@ -86,7 +103,7 @@ class TestTagFTSIndexBuild:
        fts.build_index()

        assert fts.is_ready() is True
-        assert fts.get_indexed_count() == 10
+        assert fts.get_indexed_count() == 24

    def test_build_index_nonexistent_csv(self, temp_db_path):
        """Test that build_index handles missing CSV gracefully."""
@@ -187,6 +204,110 @@ class TestTagFTSIndexSearch:
        results = populated_fts.search("girl", limit=1)
        assert len(results) <= 1

+    def test_search_tag_name_prefix_match_priority(self, populated_fts):
+        """Test that tag_name prefix matches rank higher than alias matches."""
+        results = populated_fts.search("1", limit=20)
+
+        assert len(results) > 0, "Should return results for '1'"
+
+        # Find first alias match (if any)
+        first_alias_idx = None
+        for i, result in enumerate(results):
+            if result.get("matched_alias"):
+                first_alias_idx = i
+                break
+
+        # All tag_name prefix matches should come before alias matches
+        if first_alias_idx is not None:
+            for i in range(first_alias_idx):
+                assert results[i]["tag_name"].lower().startswith("1"), (
+                    f"Tag at index {i} should start with '1' before alias matches"
+                )
+
+    def test_search_ranks_popular_tags_higher(self, populated_fts):
+        """Test that tags with higher post_count rank higher among prefix matches."""
+        results = populated_fts.search("1", limit=20)
+
+        # Filter to only tag_name prefix matches
+        prefix_matches = [r for r in results if r["tag_name"].lower().startswith("1")]
+
+        assert len(prefix_matches) > 1, "Should have multiple prefix matches"
+
+        # Verify descending post_count order among prefix matches
+        for i in range(len(prefix_matches) - 1):
+            assert (
+                prefix_matches[i]["post_count"] >= prefix_matches[i + 1]["post_count"]
+            ), (
+                f"Tags should be sorted by post_count: {prefix_matches[i]['tag_name']} ({prefix_matches[i]['post_count']}) >= {prefix_matches[i + 1]['tag_name']} ({prefix_matches[i + 1]['post_count']})"
+            )
+
+    def test_search_pagination_ordering_consistency(self, populated_fts):
+        """Test that pagination maintains consistent ordering by post_count."""
+        page1 = populated_fts.search("1", limit=10, offset=0)
+        page2 = populated_fts.search("1", limit=10, offset=10)
+
+        assert len(page1) > 0, "Page 1 should have results"
+        assert len(page2) > 0, "Page 2 should have results"
+
+        # Page 2 max post_count should be <= Page 1 min post_count
+        page1_min_posts = min(r["post_count"] for r in page1)
+        page2_max_posts = max(r["post_count"] for r in page2)
+
+        assert page2_max_posts <= page1_min_posts, (
+            f"Page 2 max post_count ({page2_max_posts}) should be <= Page 1 min post_count ({page1_min_posts})"
+        )
+
+    def test_search_returns_popular_tags_higher(self, populated_fts):
+        """Test that search returns popular tags (higher post_count) first."""
+        results = populated_fts.search("1", limit=5)
+
+        assert len(results) >= 2, "Need at least 2 results to compare"
+
+        # 1girl has 6M posts, should be ranked first
+        girl_result = next((r for r in results if r["tag_name"] == "1girl"), None)
+        assert girl_result is not None, "1girl should be in results"
+        assert results[0]["tag_name"] == "1girl", (
+            "1girl should be first due to highest post_count"
+        )
+
+        # Find a tag with significantly fewer posts
+        low_post_result = next((r for r in results if r["post_count"] < 10000), None)
+        if low_post_result:
+            assert girl_result["post_count"] > low_post_result["post_count"], (
+                f"1girl (6M posts) should have higher post_count than {low_post_result['tag_name']} ({low_post_result['post_count']} posts)"
+            )
+
+    def test_search_popularity_ordering(self, populated_fts):
+        """Test that results are ordered by post_count (popularity)."""
+        results = populated_fts.search("1", limit=20)
+
+        # Get 1girl and 1boy results for comparison
+        girl_result = next((r for r in results if r["tag_name"] == "1girl"), None)
+        boy_result = next((r for r in results if r["tag_name"] == "1boy"), None)
+
+        assert girl_result is not None, "1girl should be in results"
+        assert boy_result is not None, "1boy should be in results"
+
+        # 1girl: 6M posts, 1boy: 1.4M posts
+        assert girl_result["post_count"] == 6008644, "1girl should have 6M posts"
+        assert boy_result["post_count"] == 1405457, "1boy should have 1.4M posts"
+
+        # 1girl should rank higher due to higher post_count
+        girl_rank = results.index(girl_result)
+        boy_rank = results.index(boy_result)
+        assert girl_rank < boy_rank, (
+            f"1girl should rank higher than 1boy due to higher post_count "
+            f"(girl rank: {girl_rank}, boy rank: {boy_rank})"
+        )
+
+        # Verify results are sorted by post_count descending
+        for i in range(len(results) - 1):
+            assert results[i]["post_count"] >= results[i + 1]["post_count"], (
+                f"Results should be sorted by post_count descending: "
+                f"{results[i]['tag_name']} ({results[i]['post_count']}) >= "
+                f"{results[i + 1]['tag_name']} ({results[i + 1]['post_count']})"
+            )
+

 class TestAliasSearch:
    """Tests for alias search functionality."""
@@ -204,7 +325,9 @@ class TestAliasSearch:
        results = populated_fts.search("miku")

        assert len(results) >= 1
-        hatsune_result = next((r for r in results if r["tag_name"] == "hatsune_miku"), None)
+        hatsune_result = next(
+            (r for r in results if r["tag_name"] == "hatsune_miku"), None
+        )
        assert hatsune_result is not None
        assert hatsune_result["matched_alias"] == "miku"

@@ -214,7 +337,9 @@ class TestAliasSearch:
        results = populated_fts.search("hatsune")

        assert len(results) >= 1
-        hatsune_result = next((r for r in results if r["tag_name"] == "hatsune_miku"), None)
+        hatsune_result = next(
+            (r for r in results if r["tag_name"] == "hatsune_miku"), None
+        )
        assert hatsune_result is not None
        assert "matched_alias" not in hatsune_result

@@ -301,7 +426,9 @@ class TestSlashPrefixAliases:
    @pytest.fixture
    def fts_with_slash_aliases(self, temp_db_path):
        """Create an FTS index with slash-prefixed aliases."""
-        with tempfile.NamedTemporaryFile(mode="w", suffix=".csv", delete=False, encoding="utf-8") as f:
+        with tempfile.NamedTemporaryFile(
+            mode="w", suffix=".csv", delete=False, encoding="utf-8"
+        ) as f:
            # Format: tag_name,category,post_count,aliases
            f.write('long_hair,0,4350743,"/lh,longhair,very_long_hair"\n')
            f.write('breasts,0,3439214,"/b,boobs,oppai"\n')
@@ -380,7 +507,15 @@ class TestCategoryMappings:

    def test_category_name_to_ids_complete(self):
        """Test that CATEGORY_NAME_TO_IDS includes all expected names."""
-        expected_names = ["general", "artist", "copyright", "character", "meta", "species", "lore"]
+        expected_names = [
+            "general",
+            "artist",
+            "copyright",
+            "character",
+            "meta",
+            "species",
+            "lore",
+        ]
        for name in expected_names:
            assert name in CATEGORY_NAME_TO_IDS
            assert isinstance(CATEGORY_NAME_TO_IDS[name], list)
--- a/vue-widgets/src/components/AutocompleteTextWidget.vue
+++ b/vue-widgets/src/components/AutocompleteTextWidget.vue
@@ -7,6 +7,7 @@
        :spellcheck="spellcheck ?? false"
        :class="['text-input', { 'vue-dom-mode': isVueDomMode }]"
        @input="onInput"
+        @wheel="onWheel"
      />
      <button
        v-if="showClearButton"
@@ -82,6 +83,59 @@ const onInput = () => {
  }
 }

+/**
+ * Handle mouse wheel events on the textarea.
+ * Forwards the event to the ComfyUI canvas for zooming when the textarea has no scrollbar,
+ * or handles pinch-to-zoom gestures.
+ * 
+ * Logic aligns with ComfyUI's built-in multiline widget:
+ * src/renderer/extensions/vueNodes/widgets/composables/useStringWidget.ts
+ */
+const onWheel = (event: WheelEvent) => {
+  const textarea = textareaRef.value
+  if (!textarea) return
+
+  // Track if we have a vertical scrollbar
+  const canScrollY = textarea.scrollHeight > textarea.clientHeight
+  const deltaX = event.deltaX
+  const deltaY = event.deltaY
+  const isHorizontal = Math.abs(deltaX) > Math.abs(deltaY)
+  
+  // Access ComfyUI app from global window
+  const app = (window as any).app
+  if (!app || !app.canvas || typeof app.canvas.processMouseWheel !== 'function') {
+    return
+  }
+
+  // 1. Handle pinch-to-zoom (ctrlKey is true for pinch-to-zoom on most browsers)
+  if (event.ctrlKey) {
+    event.preventDefault()
+    event.stopPropagation()
+    app.canvas.processMouseWheel(event)
+    return
+  }
+
+  // 2. Horizontal scroll: pass to canvas (textareas usually don't scroll horizontally)
+  if (isHorizontal) {
+    event.preventDefault()
+    event.stopPropagation()
+    app.canvas.processMouseWheel(event)
+    return
+  }
+
+  // 3. Vertical scrolling:
+  if (canScrollY) {
+    // If the textarea is scrollable, let it handle the wheel event but stop propagation
+    // to prevent the canvas from zooming while the user is trying to scroll the text
+    event.stopPropagation()
+  } else {
+    // If the textarea is NOT scrollable, forward the wheel event to the canvas
+    // so it can trigger zoom in/out
+    event.preventDefault()
+    app.canvas.processMouseWheel(event)
+  }
+}
+
 // Handle external value changes (e.g., from "send lora to workflow")
 const onExternalValueChange = (event: CustomEvent<{ value: string }>) => {
  updateHasTextState()
@@ -191,7 +245,7 @@ onUnmounted(() => {
  color: var(--input-text, #ddd);
  overflow: hidden;
  overflow-y: auto;
-  padding: 2px;
+  padding: 2px 2px 24px 2px;  /* Reserve bottom space for clear button */
  resize: none;
  border: none;
  border-radius: 0;
@@ -204,7 +258,7 @@ onUnmounted(() => {
 .text-input.vue-dom-mode {
  background-color: var(--color-charcoal-400, #313235);
  color: #fff;
-  padding: 8px 12px;
+  padding: 8px 12px 30px 12px;  /* Reserve bottom space for clear button */
  margin: 0 0 4px;
  border-radius: 8px;
  font-size: 12px;
@@ -218,8 +272,8 @@ onUnmounted(() => {
 /* Clear button styles */
 .clear-button {
  position: absolute;
-  right: 4px;
-  top: 4px;
+  right: 6px;
+  bottom: 6px;  /* Changed from top to bottom */
  width: 18px;
  height: 18px;
  padding: 0;
@@ -232,11 +286,18 @@ onUnmounted(() => {
  display: flex;
  align-items: center;
  justify-content: center;
-  opacity: 0.7;
+  opacity: 0;  /* Hidden by default */
+  pointer-events: none;  /* Not clickable when hidden */
  transition: opacity 0.2s ease, background-color 0.2s ease;
  z-index: 10;
 }

+/* Show clear button when hovering over input wrapper */
+.input-wrapper:hover .clear-button {
+  opacity: 0.7;
+  pointer-events: auto;
+}
+
 .clear-button:hover {
  opacity: 1;
  background: rgba(255, 100, 100, 0.8);
@@ -250,7 +311,7 @@ onUnmounted(() => {
 /* Vue DOM mode adjustments for clear button */
 .text-input.vue-dom-mode ~ .clear-button {
  right: 8px;
-  top: 8px;
+  bottom: 10px;  /* Changed from top to bottom, adjusted for Vue DOM padding */
  width: 20px;
  height: 20px;
  background: rgba(107, 114, 128, 0.6);
--- a/web/comfyui/autocomplete.js
+++ b/web/comfyui/autocomplete.js
--- a/web/comfyui/settings.js
+++ b/web/comfyui/settings.js
@@ -19,10 +19,63 @@ const TAG_SPACE_REPLACEMENT_DEFAULT = false;
 const USAGE_STATISTICS_SETTING_ID = "loramanager.usage_statistics";
 const USAGE_STATISTICS_DEFAULT = true;

+const NEW_TAB_TEMPLATE_ID = "loramanager.new_tab_template";
+const NEW_TAB_TEMPLATE_DEFAULT = "Default";
+
+const NEW_TAB_ZOOM_LEVEL = 0.8;
+
 // ============================================================================
 // Helper Functions
 // ============================================================================

+let workflowOptions = [NEW_TAB_TEMPLATE_DEFAULT];
+let workflowOptionsFull = [{ value: "Default", label: "Default (Blank)", path: null }];
+let workflowOptionsLoaded = false;
+
+const loadWorkflowOptions = async () => {
+    if (workflowOptionsLoaded) {
+        return;
+    }
+    try {
+        const response = await fetch("/api/lm/example-workflows");
+        const data = await response.json();
+        if (data.success && data.workflows) {
+            workflowOptionsFull = data.workflows;
+            workflowOptions = data.workflows.map((w) => w.label);
+            workflowOptionsLoaded = true;
+        }
+    } catch (error) {
+        console.warn("LoRA Manager: Failed to fetch workflow options", error);
+    }
+};
+
+const getWorkflowOptions = () => {
+    // Function may be called with or without parameters
+    // Return the current workflow options array
+    return workflowOptions;
+};
+
+const loadTemplateWorkflow = async (templateName) => {
+    if (!templateName || templateName === NEW_TAB_TEMPLATE_DEFAULT) {
+        return null;
+    }
+    try {
+        const workflow = workflowOptionsFull.find((w) => w.label === templateName);
+        if (workflow && workflow.value) {
+            const workflowResponse = await fetch(
+                `/api/lm/example-workflows/${encodeURIComponent(workflow.value)}`
+            );
+            const workflowData = await workflowResponse.json();
+            if (workflowData.success && workflowData.workflow) {
+                return workflowData.workflow;
+            }
+        }
+    } catch (error) {
+        console.error("LoRA Manager: Failed to load template workflow", error);
+    }
+    return null;
+};
+
 const getWheelSensitivity = (() => {
    let settingsUnavailableLogged = false;

@@ -153,6 +206,32 @@ const getUsageStatisticsPreference = (() => {
    };
 })();

+const getNewTabTemplatePreference = (() => {
+    let settingsUnavailableLogged = false;
+
+    return () => {
+        const settingManager = app?.extensionManager?.setting;
+        if (!settingManager || typeof settingManager.get !== "function") {
+            if (!settingsUnavailableLogged) {
+                console.warn("LoRA Manager: settings API unavailable, using default new tab template.");
+                settingsUnavailableLogged = true;
+            }
+            return NEW_TAB_TEMPLATE_DEFAULT;
+        }
+
+        try {
+            const value = settingManager.get(NEW_TAB_TEMPLATE_ID);
+            return value ?? NEW_TAB_TEMPLATE_DEFAULT;
+        } catch (error) {
+            if (!settingsUnavailableLogged) {
+                console.warn("LoRA Manager: unable to read new tab template setting, using default.", error);
+                settingsUnavailableLogged = true;
+            }
+            return NEW_TAB_TEMPLATE_DEFAULT;
+        }
+    };
+})();
+
 // ============================================================================
 // Register Extension with All Settings
 // ============================================================================
@@ -205,11 +284,95 @@ app.registerExtension({
            tooltip: "When enabled, LoRA Manager will track model usage statistics during workflow execution. Disabling this will prevent unnecessary disk writes.",
            category: ["LoRA Manager", "Statistics", "Usage Tracking"],
        },
+        {
+            id: NEW_TAB_TEMPLATE_ID,
+            name: "New Tab Template Workflow",
+            type: "combo",
+            options: getWorkflowOptions,
+            defaultValue: NEW_TAB_TEMPLATE_DEFAULT,
+            tooltip: "Choose a template workflow to load when creating a new workflow tab. 'Default (Blank)' keeps ComfyUI's original blank workflow behavior.",
+            category: ["LoRA Manager", "Workflow", "New Tab Template"],
+        },
    ],
+    async setup() {
+        await loadWorkflowOptions();
+
+        const originalNewBlankWorkflow = async () => {
+            const blankGraph = {
+                last_node_id: 0,
+                last_link_id: 0,
+                nodes: [],
+                links: [],
+                groups: [],
+                config: {},
+                extra: {},
+                version: 0.4,
+            };
+            await app.loadGraphData(blankGraph);
+        };
+
+        const waitForCommandStore = async (maxWaitMs = 5000) => {
+            const startTime = Date.now();
+            while (Date.now() - startTime < maxWaitMs) {
+                if (app.extensionManager?.command?.commands) {
+                    return true;
+                }
+                await new Promise((resolve) => setTimeout(resolve, 100));
+            }
+            return false;
+        };
+
+        const patchCommand = async () => {
+            const storeReady = await waitForCommandStore();
+            if (!storeReady) {
+                console.warn("LoRA Manager: Could not access command store to patch NewBlankWorkflow");
+                return;
+            }
+
+            const commands = app.extensionManager.command.commands;
+            for (const cmd of commands) {
+                if (cmd.id === "Comfy.NewBlankWorkflow") {
+                    const originalFunc = cmd.function;
+                    cmd.function = async (metadata) => {
+                        const templateName = getNewTabTemplatePreference();
+                        
+                        if (templateName && templateName !== NEW_TAB_TEMPLATE_DEFAULT) {
+                            const workflowData = await loadTemplateWorkflow(templateName);
+                            if (workflowData) {
+                                // Override the workflow's saved view settings with our custom zoom
+                                if (!workflowData.extra) {
+                                    workflowData.extra = {};
+                                }
+                                if (!workflowData.extra.ds) {
+                                    workflowData.extra.ds = { offset: [0, 0], scale: 1 };
+                                }
+                                workflowData.extra.ds.scale = NEW_TAB_ZOOM_LEVEL;
+                                
+                                await app.loadGraphData(workflowData);
+                                return;
+                            }
+                        }
+                        
+                        await originalNewBlankWorkflow();
+                    };
+                    break;
+                }
+            }
+        };
+
+        patchCommand();
+    },
 });

 // ============================================================================
 // Exports
 // ============================================================================

-export { getWheelSensitivity, getAutoPathCorrectionPreference, getPromptTagAutocompletePreference, getTagSpaceReplacementPreference, getUsageStatisticsPreference };
+export {
+    getWheelSensitivity,
+    getAutoPathCorrectionPreference,
+    getPromptTagAutocompletePreference,
+    getTagSpaceReplacementPreference,
+    getUsageStatisticsPreference,
+    getNewTabTemplatePreference,
+};
--- a/web/comfyui/top_menu_extension.js
+++ b/web/comfyui/top_menu_extension.js
@@ -3,6 +3,10 @@ import { app } from "../../scripts/app.js";
 const BUTTON_TOOLTIP = "Launch LoRA Manager (Shift+Click opens in new window)";
 const LORA_MANAGER_PATH = "/loras";
 const NEW_WINDOW_FEATURES = "width=1200,height=800,resizable=yes,scrollbars=yes,status=yes";
+const MAX_ATTACH_ATTEMPTS = 120;
+const BUTTON_GROUP_CLASS = "lora-manager-top-menu-group";
+
+const MIN_VERSION_FOR_ACTION_BAR = [1, 33, 9];

 const openLoraManager = (event) => {
    const url = `${window.location.origin}${LORA_MANAGER_PATH}`;
@@ -15,6 +19,65 @@ const openLoraManager = (event) => {
    window.open(url, "_blank");
 };

+const getComfyUIFrontendVersion = async () => {
+    try {
+        if (window['__COMFYUI_FRONTEND_VERSION__']) {
+            return window['__COMFYUI_FRONTEND_VERSION__'];
+        }
+    } catch (error) {
+        console.warn("LoRA Manager: unable to read __COMFYUI_FRONTEND_VERSION__:", error);
+    }
+
+    try {
+        const response = await fetch("/system_stats");
+        const data = await response.json();
+        
+        if (data?.system?.comfyui_frontend_version) {
+            return data.system.comfyui_frontend_version;
+        }
+        
+        if (data?.system?.required_frontend_version) {
+            return data.system.required_frontend_version;
+        }
+    } catch (error) {
+        console.warn("LoRA Manager: unable to fetch system_stats:", error);
+    }
+
+    return "0.0.0";
+};
+
+const parseVersion = (versionStr) => {
+    if (!versionStr || typeof versionStr !== 'string') {
+        return [0, 0, 0];
+    }
+    
+    const cleanVersion = versionStr.replace(/^[vV]/, '').split('-')[0];
+    const parts = cleanVersion.split('.').map(part => parseInt(part, 10) || 0);
+    
+    while (parts.length < 3) {
+        parts.push(0);
+    }
+    
+    return parts;
+};
+
+const compareVersions = (version1, version2) => {
+    const v1 = typeof version1 === 'string' ? parseVersion(version1) : version1;
+    const v2 = typeof version2 === 'string' ? parseVersion(version2) : version2;
+    
+    for (let i = 0; i < 3; i++) {
+        if (v1[i] > v2[i]) return 1;
+        if (v1[i] < v2[i]) return -1;
+    }
+    
+    return 0;
+};
+
+const supportsActionBarButtons = async () => {
+    const version = await getComfyUIFrontendVersion();
+    return compareVersions(version, MIN_VERSION_FOR_ACTION_BAR) >= 0;
+};
+
 const fetchVersionInfo = async () => {
    try {
        const response = await fetch("/api/lm/version-info");
@@ -30,6 +93,55 @@ const fetchVersionInfo = async () => {
    return "";
 };

+const createTopMenuButton = async () => {
+    const { ComfyButton } = await import("../../scripts/ui/components/button.js");
+    
+    const button = new ComfyButton({
+        icon: "loramanager",
+        tooltip: BUTTON_TOOLTIP,
+        app,
+        enabled: true,
+        classList: "comfyui-button comfyui-menu-mobile-collapse primary",
+    });
+
+    button.element.setAttribute("aria-label", BUTTON_TOOLTIP);
+    button.element.title = BUTTON_TOOLTIP;
+
+    if (button.iconElement) {
+        button.iconElement.innerHTML = getLoraManagerIcon();
+        button.iconElement.style.width = "1.2rem";
+        button.iconElement.style.height = "1.2rem";
+    }
+
+    button.element.addEventListener("click", openLoraManager);
+    return button;
+};
+
+const attachTopMenuButton = async (attempt = 0) => {
+    if (document.querySelector(`.${BUTTON_GROUP_CLASS}`)) {
+        return;
+    }
+
+    const settingsGroup = app.menu?.settingsGroup;
+    if (!settingsGroup?.element?.parentElement) {
+        if (attempt >= MAX_ATTACH_ATTEMPTS) {
+            console.warn("LoRA Manager: unable to locate the ComfyUI settings button group.");
+            return;
+        }
+
+        requestAnimationFrame(() => attachTopMenuButton(attempt + 1));
+        return;
+    }
+
+    const loraManagerButton = await createTopMenuButton();
+    const { ComfyButtonGroup } = await import("../../scripts/ui/components/buttonGroup.js");
+    
+    const buttonGroup = new ComfyButtonGroup(loraManagerButton);
+    buttonGroup.element.classList.add(BUTTON_GROUP_CLASS);
+
+    settingsGroup.element.before(buttonGroup.element);
+};
+
 const createAboutBadge = (version) => {
    const label = version ? `LoRA Manager v${version}` : "LoRA Manager";

@@ -40,60 +152,80 @@ const createAboutBadge = (version) => {
    };
 };

-app.registerExtension({
-    name: "LoraManager.TopMenu",
-    actionBarButtons: [
-        {
-            icon: "icon-[mdi--alpha-l-box] size-4",
-            tooltip: BUTTON_TOOLTIP,
-            onClick: openLoraManager
-        }
-    ],
-    aboutPageBadges: [createAboutBadge()],
-    async setup() {
-        const version = await fetchVersionInfo();
-        this.aboutPageBadges = [createAboutBadge(version)];
-
-        const injectStyles = () => {
-            const styleId = 'lm-top-menu-button-styles';
-            if (document.getElementById(styleId)) return;
-
-            const style = document.createElement('style');
-            style.id = styleId;
-            style.textContent = `
-                button[aria-label="Launch LoRA Manager (Shift+Click opens in new window)"].lm-top-menu-button {
-                    transition: all 0.2s ease;
-                    border: 1px solid transparent;
-                }
-                button[aria-label="Launch LoRA Manager (Shift+Click opens in new window)"].lm-top-menu-button:hover {
-                    background-color: var(--primary-hover-bg) !important;
-                }
-            `;
-            document.head.appendChild(style);
-        };
-        injectStyles();
-
-        const replaceButtonIcon = () => {
-            const buttons = document.querySelectorAll('button[aria-label="Launch LoRA Manager (Shift+Click opens in new window)"]');
-            buttons.forEach(button => {
-                button.classList.add('lm-top-menu-button');
-                button.innerHTML = getLoraManagerIcon();
-                button.style.borderRadius = '4px';
-                button.style.padding = '6px';
-                button.style.backgroundColor = 'var(--primary-bg)';
-                const svg = button.querySelector('svg');
-                if (svg) {
-                    svg.style.width = '20px';
-                    svg.style.height = '20px';
-                }
-            });
-            if (buttons.length === 0) {
-                requestAnimationFrame(replaceButtonIcon);
+const createExtensionObject = (useActionBar) => {
+    const extensionObj = {
+        name: "LoraManager.TopMenu",
+        async setup() {
+            const version = await fetchVersionInfo();
+            
+            if (!useActionBar) {
+                console.log("LoRA Manager: using legacy button attachment (frontend version < 1.33.9)");
+                await attachTopMenuButton();
+            } else {
+                console.log("LoRA Manager: using actionBarButtons API (frontend version >= 1.33.9)");
            }
-        };
-        requestAnimationFrame(replaceButtonIcon);
-    },
-});
+            
+            this.aboutPageBadges = [createAboutBadge(version)];
+
+            const injectStyles = () => {
+                const styleId = 'lm-top-menu-button-styles';
+                if (document.getElementById(styleId)) return;
+
+                const style = document.createElement('style');
+                style.id = styleId;
+                style.textContent = `
+                    button[aria-label="Launch LoRA Manager (Shift+Click opens in new window)"].lm-top-menu-button {
+                        transition: all 0.2s ease;
+                        border: 1px solid transparent;
+                    }
+                    button[aria-label="Launch LoRA Manager (Shift+Click opens in new window)"].lm-top-menu-button:hover {
+                        background-color: var(--primary-hover-bg) !important;
+                    }
+                `;
+                document.head.appendChild(style);
+            };
+            injectStyles();
+
+            const replaceButtonIcon = () => {
+                const buttons = document.querySelectorAll('button[aria-label="Launch LoRA Manager (Shift+Click opens in new window)"]');
+                buttons.forEach(button => {
+                    button.classList.add('lm-top-menu-button');
+                    button.innerHTML = getLoraManagerIcon();
+                    button.style.borderRadius = '4px';
+                    button.style.padding = '6px';
+                    button.style.backgroundColor = 'var(--primary-bg)';
+                    const svg = button.querySelector('svg');
+                    if (svg) {
+                        svg.style.width = '20px';
+                        svg.style.height = '20px';
+                    }
+                });
+                if (buttons.length === 0) {
+                    requestAnimationFrame(replaceButtonIcon);
+                }
+            };
+            requestAnimationFrame(replaceButtonIcon);
+        },
+    };
+    
+    if (useActionBar) {
+        extensionObj.actionBarButtons = [
+            {
+                icon: "icon-[mdi--alpha-l-box] size-4",
+                tooltip: BUTTON_TOOLTIP,
+                onClick: openLoraManager
+            }
+        ];
+    }
+    
+    return extensionObj;
+};
+
+(async () => {
+    const useActionBar = await supportsActionBarButtons();
+    const extensionObj = createExtensionObject(useActionBar);
+    app.registerExtension(extensionObj);
+})();

 const getLoraManagerIcon = () => {
    return `
--- a/web/comfyui/vue-widgets/lora-manager-widgets.js
+++ b/web/comfyui/vue-widgets/lora-manager-widgets.js
@@ -1988,14 +1988,14 @@ to { transform: rotate(360deg);
  padding: 20px 0;
 }

-.autocomplete-text-widget[data-v-653446fd] {
+.autocomplete-text-widget[data-v-b3b00fdd] {
  background: transparent;
  height: 100%;
  display: flex;
  flex-direction: column;
  box-sizing: border-box;
 }
-.input-wrapper[data-v-653446fd] {
+.input-wrapper[data-v-b3b00fdd] {
  position: relative;
  flex: 1;
  display: flex;
@@ -2003,14 +2003,14 @@ to { transform: rotate(360deg);
 }

 /* Canvas mode styles (default) - matches built-in comfy-multiline-input */
-.text-input[data-v-653446fd] {
+.text-input[data-v-b3b00fdd] {
  flex: 1;
  width: 100%;
  background-color: var(--comfy-input-bg, #222);
  color: var(--input-text, #ddd);
  overflow: hidden;
  overflow-y: auto;
-  padding: 2px;
+  padding: 2px 2px 24px 2px;  /* Reserve bottom space for clear button */
  resize: none;
  border: none;
  border-radius: 0;
@@ -2020,24 +2020,24 @@ to { transform: rotate(360deg);
 }

 /* Vue DOM mode styles - matches built-in p-textarea in Vue DOM mode */
-.text-input.vue-dom-mode[data-v-653446fd] {
+.text-input.vue-dom-mode[data-v-b3b00fdd] {
  background-color: var(--color-charcoal-400, #313235);
  color: #fff;
-  padding: 8px 12px;
+  padding: 8px 12px 30px 12px;  /* Reserve bottom space for clear button */
  margin: 0 0 4px;
  border-radius: 8px;
  font-size: 12px;
  font-family: inherit;
 }
-.text-input[data-v-653446fd]:focus {
+.text-input[data-v-b3b00fdd]:focus {
  outline: none;
 }

 /* Clear button styles */
-.clear-button[data-v-653446fd] {
+.clear-button[data-v-b3b00fdd] {
  position: absolute;
-  right: 4px;
-  top: 4px;
+  right: 6px;
+  bottom: 6px;  /* Changed from top to bottom */
  width: 18px;
  height: 18px;
  padding: 0;
@@ -2050,31 +2050,38 @@ to { transform: rotate(360deg);
  display: flex;
  align-items: center;
  justify-content: center;
-  opacity: 0.7;
+  opacity: 0;  /* Hidden by default */
+  pointer-events: none;  /* Not clickable when hidden */
  transition: opacity 0.2s ease, background-color 0.2s ease;
  z-index: 10;
 }
-.clear-button[data-v-653446fd]:hover {
+
+/* Show clear button when hovering over input wrapper */
+.input-wrapper:hover .clear-button[data-v-b3b00fdd] {
+  opacity: 0.7;
+  pointer-events: auto;
+}
+.clear-button[data-v-b3b00fdd]:hover {
  opacity: 1;
  background: rgba(255, 100, 100, 0.8);
 }
-.clear-button svg[data-v-653446fd] {
+.clear-button svg[data-v-b3b00fdd] {
  width: 12px;
  height: 12px;
 }

 /* Vue DOM mode adjustments for clear button */
-.text-input.vue-dom-mode ~ .clear-button[data-v-653446fd] {
+.text-input.vue-dom-mode ~ .clear-button[data-v-b3b00fdd] {
  right: 8px;
-  top: 8px;
+  bottom: 10px;  /* Changed from top to bottom, adjusted for Vue DOM padding */
  width: 20px;
  height: 20px;
  background: rgba(107, 114, 128, 0.6);
 }
-.text-input.vue-dom-mode ~ .clear-button[data-v-653446fd]:hover {
+.text-input.vue-dom-mode ~ .clear-button[data-v-b3b00fdd]:hover {
  background: oklch(62% 0.18 25);
 }
-.text-input.vue-dom-mode ~ .clear-button svg[data-v-653446fd] {
+.text-input.vue-dom-mode ~ .clear-button svg[data-v-b3b00fdd] {
  width: 14px;
  height: 14px;
 }`));
@@ -14232,6 +14239,36 @@ const _sfc_main = /* @__PURE__ */ defineComponent({
        props.widget.callback(textareaRef.value.value);
      }
    };
+    const onWheel = (event) => {
+      const textarea = textareaRef.value;
+      if (!textarea) return;
+      const canScrollY = textarea.scrollHeight > textarea.clientHeight;
+      const deltaX = event.deltaX;
+      const deltaY = event.deltaY;
+      const isHorizontal = Math.abs(deltaX) > Math.abs(deltaY);
+      const app2 = window.app;
+      if (!app2 || !app2.canvas || typeof app2.canvas.processMouseWheel !== "function") {
+        return;
+      }
+      if (event.ctrlKey) {
+        event.preventDefault();
+        event.stopPropagation();
+        app2.canvas.processMouseWheel(event);
+        return;
+      }
+      if (isHorizontal) {
+        event.preventDefault();
+        event.stopPropagation();
+        app2.canvas.processMouseWheel(event);
+        return;
+      }
+      if (canScrollY) {
+        event.stopPropagation();
+      } else {
+        event.preventDefault();
+        app2.canvas.processMouseWheel(event);
+      }
+    };
    const onExternalValueChange = (event) => {
      updateHasTextState();
    };
@@ -14290,7 +14327,8 @@ const _sfc_main = /* @__PURE__ */ defineComponent({
            placeholder: __props.placeholder,
            spellcheck: __props.spellcheck ?? false,
            class: normalizeClass(["text-input", { "vue-dom-mode": isVueDomMode.value }]),
-            onInput
+            onInput,
+            onWheel
          }, null, 42, _hoisted_3),
          showClearButton.value ? (openBlock(), createElementBlock("button", {
            key: 0,
@@ -14324,7 +14362,7 @@ const _sfc_main = /* @__PURE__ */ defineComponent({
    };
  }
 });
-const AutocompleteTextWidget = /* @__PURE__ */ _export_sfc(_sfc_main, [["__scopeId", "data-v-653446fd"]]);
+const AutocompleteTextWidget = /* @__PURE__ */ _export_sfc(_sfc_main, [["__scopeId", "data-v-b3b00fdd"]]);
 const LORA_PROVIDER_NODE_TYPES$1 = [
  "Lora Stacker (LoraManager)",
  "Lora Randomizer (LoraManager)",
--- a/web/comfyui/vue-widgets/lora-manager-widgets.js.map
+++ b/web/comfyui/vue-widgets/lora-manager-widgets.js.map
Author	SHA1	Message	Date
Will Miao	2dae4c1291	fix: isolate extra unet paths from checkpoints to prevent type misclassification Refactor _prepare_checkpoint_paths() to return a tuple instead of having side effects on instance variables. This prevents extra unet paths from being incorrectly classified as checkpoints when processing extra paths. - Changed return type from List[str] to Tuple[List[str], List[str], List[str]] (all_paths, checkpoint_roots, unet_roots) - Updated _init_checkpoint_paths() and _apply_library_paths() callers - Fixed extra paths processing to properly isolate main and extra roots - Updated test_checkpoint_path_overlap.py tests for new API This ensures models in extra unet paths are correctly identified as diffusion_model type and don't appear in checkpoints list.	2026-03-17 22:03:57 +08:00
Will Miao	70c150bd80	fix(services): implement stable sorting for model and recipe caches Add file_path as a tie-breaker for all sort modes in ModelCache, BaseModelService, LoraService, and RecipeCache to ensure deterministic ordering when primary keys are identical. Resolves issue #859.	2026-03-17 14:20:23 +08:00
Will Miao	9e81c33f8a	fix(utils): make sanitize_folder_name idempotent by combining strip/rstrip calls	2026-03-17 11:24:59 +08:00
Will Miao	22c0dbd734	feat(recipes): persist 'Skip images without metadata' choice in batch import	2026-03-17 11:01:41 +08:00
Will Miao	d0c58472be	fix(i18n): add missing common.actions.close translation key	2026-03-17 09:57:27 +08:00
Will Miao	b3c530bf36	fix(autocomplete): handle multi-word tag matching with normalized spaces - Replace multiple consecutive spaces with single underscore for tag matching (e.g., 'looking to the side' → 'looking_to_the_side') - Support prefix/suffix matching for flexible multi-word autocomplete (e.g., 'looking to the' → 'looking_to_the_side') - Add comprehensive test coverage for multi-word scenarios Test coverage: - Multi-word exact match (Danbooru convention) - Partial match with last token replacement - Command mode with multi-word phrases - Multiple consecutive spaces handling - Backend LOG10 popularity weight validation Fixes: 'looking to the side' input now correctly replaces with 'looking_to_the_side, ' (or 'looking to the side, ' with space replacement)	2026-03-17 09:34:01 +08:00
Will Miao	b5a0725d2c	fix(autocomplete): improve tag search ranking with popularity-based sorting - Add LOG10(post_count) weighting to BM25 score for better relevance ranking - Prioritize tag_name prefix matches above alias matches using CASE statement - Remove frontend re-scoring logic to trust backend排序 results - Fix pagination consistency: page N+1 scores <= page N minimum score Key improvements: - '1girl' (6M posts) now ranks #1 instead of #149 for search '1' - tag_name prefix matches always appear before alias matches - Popular tags rank higher than obscure ones with same prefix - Consistent ordering across pagination boundaries Test coverage: - Add test_search_tag_name_prefix_match_priority - Add test_search_ranks_popular_tags_higher - Add test_search_pagination_ordering_consistency - Add test_search_rank_score_includes_popularity_weight - Update test data with 15 tags starting with '1' Fixes issues with autocomplete dropdown showing inconsistent results when scrolling through paginated search results.	2026-03-16 19:09:07 +08:00
Will Miao	ef38bda04f	docs: remove redundant example metadata files (#856 ) - Delete examples/metadata/ directory and all example files - Real metadata.json files in model roots are better examples - Examples were artificial and could become outdated - Maintenance burden outweighs benefit - Remove 'Complete Examples' section from docs/metadata-json-schema.md - Remove reference to example files in 'See Also' section Rationale: Users have access to real-world metadata.json files in their actual model directories, which contain complete Civitai API responses with authentic data structures (images arrays with prompts, files with hashes, creator information, etc.). These are more valuable than simplified artificial examples.	2026-03-16 09:41:58 +08:00
Will Miao	58713ea6e0	fix(top-menu): use dynamic imports to eliminate deprecation warnings - Replace static imports of deprecated ComfyButton and ComfyButtonGroup with dynamic imports - Only loads legacy API files when frontend version < 1.33.9 (backward compatibility path) - Frontend >= 1.33.9 users no longer see deprecation warnings since legacy code is never loaded - Preserves full backward compatibility for older ComfyUI frontend versions - All existing tests pass (159 JS + 65 Vue tests)	2026-03-16 09:41:58 +08:00
Will Miao	8b91920058	docs: add comprehensive metadata.json schema documentation (#856 ) - Create docs/metadata-json-schema.md with complete field reference - All base fields for LoRA, Checkpoint, and Embedding models - Complete civitai object structure with Used vs Stored field classification - Model-level fields (allowCommercialUse, allowDerivatives, etc.) - Creator fields (username, image) - customImages structure with actual field names and types - Field behavior categories (Auto-Updated, Set Once, User-Editable) - Add .specs/metadata.schema.json for programmatic validation - JSON Schema draft-07 format - oneOf schemas for each model type - Definitions for civitaiObject and usageTips - Add example metadata files for each model type - lora-civitai.json: LoRA with full Civitai data - lora-custom.json: User-defined LoRA with trigger words - lora-no-triggerwords.json: LoRA without trigger words - checkpoint-civitai.json: Checkpoint from Civitai - embedding-custom.json: Custom embedding Key clarifications: - modified: Import timestamp (Set Once, never changes after import) - size: File size at import time (Set Once) - base_model: Optional with actual values (SDXL 1.0, Flux.1 D, etc.) - model_type: Used in metadata.json (not sub_type which is internal) - allowCommercialUse: ["Image", "Video", "RentCivit", "Rent"] - civitai.files/images: Marked as Used by Lora Manager - User-editable fields clearly documented (model_name, tags, etc.)	2026-03-16 09:41:58 +08:00
Will Miao	ee466113d5	feat: implement batch import recipe functionality (frontend + backend fixes) Backend fixes: - Add missing API route for /api/lm/recipes/batch-import/progress (GET) - Add missing API route for /api/lm/recipes/batch-import/directory (POST) - Add missing API route for /api/lm/recipes/browse-directory (POST) - Register WebSocket endpoint for batch import progress - Fix skip_no_metadata default value (True -> False) to allow no-LoRA imports - Add items array to BatchImportProgress.to_dict() for detailed results Frontend implementation: - Create BatchImportManager.js with complete batch import workflow - Add directory browser UI for selecting folders - Add batch import modal with URL list and directory input modes - Implement real-time progress tracking (WebSocket + HTTP polling) - Add results summary with success/failed/skipped statistics - Add expandable details view showing individual item status - Auto-refresh recipe list after import completion UI improvements: - Add spinner animation for importing status - Simplify results summary UI to match progress stats styling - Fix current item text alignment - Fix dark theme styling for directory browser button - Fix batch import button styling consistency Translations: - Add batch import related i18n keys to all locale files - Run sync_translation_keys.py to sync all translations Fixes: - Batch import now allows images without LoRAs (matches single import behavior) - Progress endpoint now returns complete items array with status details - Results view correctly displays skipped items with error messages	2026-03-16 09:41:58 +08:00
Will Miao	f86651652c	feat(batch-import): implement backend batch import service with adaptive concurrency - Add BatchImportService with concurrent execution using asyncio.gather - Implement AdaptiveConcurrencyController with dynamic adjustment - Add input validation for URLs and local paths - Support duplicate detection via skip_duplicates parameter - Add WebSocket progress broadcasting for real-time updates - Create comprehensive unit tests for batch import functionality - Update API handlers and route registrations - Add i18n translation keys for batch import UI	2026-03-16 09:41:58 +08:00
Will Miao	c89d4dae85	fix(extra-paths): support trigger words for LoRAs in extra folder paths, fixes #860 - Update get_lora_info() to check both loras_roots and extra_loras_roots - Add fallback logic to return trigger words even if path not in recognized roots - Ensure Trigger Word Toggle node displays trigger words for LoRAs from extra folder paths Fixes issue where LoRAs added from extra folder paths would not show their trigger words in connected Trigger Word Toggle nodes.	2026-03-16 09:38:21 +08:00
pixelpaws	55a18d401b	Merge pull request #858 from botchedchuckle/patch-1 Fix: Escape HTML in Prompt/NegativePrompt for MetadataPanel	2026-03-14 14:43:46 +08:00
botchedchuckle	7570936c75	Fix: Escape HTML in Prompt/NegativePrompt for MetadataPanel * Fixed a bug where `prompt` and `negativePrompt` were both being added directly to HTML without escaping them. Given prompts are allowed to have HTML characters (e.g. `<lora:something:0.75>`), by forgetting to escape them some tags were missing in the metadata views for example images using those characters.	2026-03-13 01:29:04 -07:00
Will Miao	4fcf641d57	fix(bulk-context-menu): escape special characters in data-filepath selector to support double quotes in filenames (#845 )	2026-03-12 08:49:10 +08:00
Will Miao	5c29e26c4e	fix(top-menu): add backward compatibility for actionBarButtons API (#853 ) - Implement version detection using __COMFYUI_FRONTEND_VERSION__ and /system_stats API - Add version parsing and comparison utilities - Dynamically register extension based on frontend version - Use actionBarButtons API for frontend >= 1.33.9 - Fallback to legacy ComfyButton approach for older versions - Add comprehensive version detection tests	2026-03-12 07:41:29 +08:00
Will Miao	ee765a6d22	fix(sidebar): escape folder names and paths to support double quotes - Import and use escapeHtml and escapeAttribute in SidebarManager.js - Escape data-path and title attributes in folder tree and breadcrumbs - Use CSS.escape() for attribute selectors in updateTreeSelection - Fixes issue #843 where folders with double quotes broke navigation	2026-03-11 23:33:11 +08:00
Will Miao	c02f603ed2	fix(autocomplete): add wheel event handler for canvas zoom support Add @wheel event listener to AutocompleteTextWidget textarea to enable canvas zoom when textarea has no scrollbar. The onWheel handler: - Forwards pinch-to-zoom (ctrl+wheel) to canvas - Passes horizontal scroll to canvas - When textarea has vertical scrollbar: lets textarea scroll - When textarea has NO scrollbar: forwards to canvas for zoom Behavior now matches ComfyUI built-in multiline widget. Fixes #850	2026-03-11 20:58:01 +08:00
Will Miao	ee84b30023	Fix node selector z-index issue in recipe modal Change node-selector z-index from 1000 to var(--z-overlay) (2000) to ensure the model selector UI appears above the recipe modal when sending checkpoints to workflow with multiple targets.	2026-03-09 19:29:13 +08:00
Will Miao	97979d9e7c	fix(send-to-workflow): strip file extension before searching relative paths Backend _relative_path_matches_tokens() removes extensions from paths before matching (commit `43f6bfab`). This fix ensures frontend also removes extensions from search terms to avoid matching failures. Fixes issue where send model to workflow would receive absolute paths instead of relative paths because the API returned empty results when searching with file extension.	2026-03-09 15:49:37 +08:00
Will Miao	cda271890a	feat(workflow-template): add new tab template workflow with auto-zoom - Add GET /api/lm/example-workflows endpoint to list available templates - Add GET /api/lm/example-workflows/{filename} to retrieve specific workflow - Add 'New Tab Template Workflow' setting in LoRA Manager settings - Automatically apply 80% zoom level when loading template workflows - Override workflow's saved view settings to prevent visual zoom flicker The feature allows users to select a template workflow from example_workflows/ directory to load when creating new workflow tabs, with a hardcoded 0.8 zoom level for better initial view experience.	2026-03-08 21:03:14 +08:00
Will Miao	2fbe6c8843	fix(autocomplete): fix dropdown width calculation bug Temporarily remove width constraints when measuring content to prevent scrollWidth from being limited by narrow container. This fixes the issue where dropdown width was incorrectly calculated as ~120px. Also update test to match maxItems default value (100).	2026-03-07 23:23:26 +08:00
Will Miao	4fb07370dd	fix(tests): add offset parameter to MockTagFTSIndex.search() Add missing offset parameter to MockTagFTSIndex to support pagination changes from commit `a802a89`. - Update search() signature to include offset=0 - Implement pagination logic with offset/limit slicing	2026-03-07 23:10:00 +08:00
Will Miao	43f6bfab36	fix(autocomplete): strip file extensions from model names in search suggestions Remove .safetensors/.ckpt/.pt/.bin extensions from model names in autocomplete suggestions to improve UX and search relevance: Frontend (web/comfyui/autocomplete.js): - Add _getDisplayText() helper to strip extensions from model paths - Update _matchItem() to match against filename without extension - Update render() and createItemElement() to display clean names Backend (py/services/base_model_service.py): - Add _remove_model_extension() helper method - Update _relative_path_matches_tokens() to ignore extensions in matching - Update _relative_path_sort_key() to sort based on names without extensions Tests (tests/services/test_relative_path_search.py): - Add tests to verify 's' and 'safe' queries don't match all .safetensors files Fixes issue where typing 's' would match all .safetensors files and cluttered suggestions with redundant extension names.	2026-03-07 23:07:10 +08:00
Will Miao	a802a89ff9	feat(autocomplete): implement virtual scrolling and pagination - Add virtual scrolling with configurable visible items (default: 15) - Implement pagination with offset/limit for backend APIs - Support loading more items on scroll - Fix width calculation for suggestions dropdown - Update backend services to support offset parameter Files modified: - web/comfyui/autocomplete.js (virtual scroll, pagination) - py/services/base_model_service.py (offset support) - py/services/custom_words_service.py (offset support) - py/services/tag_fts_index.py (offset support) - py/routes/handlers/model_handlers.py (offset param) - py/routes/handlers/misc_handlers.py (offset param)	2026-03-07 22:17:26 +08:00
Will Miao	343dd91e4b	feat(ui): improve clear button UX in autocomplete text widget Move clear button from top-right to bottom-right to avoid obscuring text content. Add hover visibility for cleaner UI. Reserve bottom padding in textarea for button placement.	2026-03-07 21:09:59 +08:00
Will Miao	3756f88368	feat(autocomplete): improve multi-word tag search with query normalization Implement search query variation generation to improve matching for multi-word tags: - Generate multiple query forms: original, underscore (spaces->_), no-space, last token - Execute up to 4 parallel queries with result merging and deduplication - Add smart matching with symbol-insensitive comparison (blue hair matches blue_hair) - Sort results with exact matches prioritized over partial matches This allows users to type natural language queries like 'looking to the side' and find tags like 'Looking_to_the_side' while maintaining backward compatibility with continuous typing workflows.	2026-03-07 20:24:35 +08:00
Will Miao	acc625ead3	feat(recipes): add sync changes dropdown menu for recipe refresh - Add syncChanges() function to recipeApi.js for quick refresh without cache rebuild - Implement dropdown menu UI in recipes page with quick refresh and full rebuild options - Add initDropdowns() method to RecipeManager for dropdown interaction handling - Update AGENTS.md with more precise instruction about running sync_translation_keys.py - Integrate sync changes functionality as default refresh behavior	2026-03-04 20:31:58 +08:00
Will Miao	f402505f97	i18n: complete TODO translations in locale files - Add missing translations for modelTypes, recipe refresh, and sync notifications - Translate for all supported languages (zh-CN, zh-TW, ja, ko, fr, de, es, ru, he) - Run sync_translation_keys.py to ensure key consistency	2026-03-04 20:27:21 +08:00
Will Miao	4d8113464c	perf(recipe_scanner): eliminate event loop blocking during cache rebuild Refactor force_refresh path to use thread pool execution instead of blocking the event loop shared with ComfyUI. Key changes: - Fix 1: Route force_refresh through _initialize_recipe_cache_sync() in thread pool - Fix 2: Add GIL release points (time.sleep(0)) every 100 files in sync loops - Fix 3: Move RecipeCache.resort() to thread pool via run_in_executor - Fix 4: Persist cache automatically after force_refresh - Fix 5: Increase yield frequency in _enrich_cache_metadata (every recipe) This eliminates the ~5 minute freeze when rebuilding 30K recipe cache. Fixes performance issue where ComfyUI became unresponsive during recipe scanning due to shared Python event loop blocking.	2026-03-04 15:10:46 +08:00
Will Miao	1ed503a6b5	docs: add lazy hash computation to v1.0.0 release notes	2026-03-04 07:41:19 +08:00
Will Miao	d67914e095	docs: update portable package download link to v1.0.0	2026-03-03 22:06:29 +08:00