CivitAI's PaidAccess cutover deprecated the availability=EarlyAccess and
earlyAccessEndsAt signals; gated versions now report availability=Public
with a paidAccess DTO that LoRA Manager previously ignored, so "Hide
Early Access Updates" missed paid/early-access models and downloads
failed with 401.
Parse and persist paidAccess from model-level, bulk, and by-hash
responses; treat timed paid gates as early access and permanent paid
versions as a distinct is_paid state; add a hide_paid_updates setting
with a "Paid" badge in the versions tab; warn before downloading gated
versions. Includes SQLite migration, i18n for all locales, and
backend/frontend tests.
Track recipe modal opens in a separate stats file (never touching recipe
JSON/EXIF), expose a fire-and-forget POST endpoint, and add an 'opened'
sort that hides never-opened recipes as a true recently-opened view.
Includes i18n for all locales and backend/frontend tests.
load_checkpoint returns a 4-tuple (MODEL, CLIP, VAE, model_name) since the
random loader exposes the selected model name; the annotation still claimed
a 3-tuple.
The sampler field now accepts either a manual string or a SAMPLER
connection. When wired, the sampler name is extracted from the
KSAMPLER object's sampler_function __name__ (sample_euler -> euler),
with special-casing for dpm_fast/dpm_adaptive local closures and
uni_pc/uni_pc_bh2 function names.
- sampler input declared as "STRING,SAMPLER" with widgetType STRING,
mirroring the existing model field union pattern
- shared collect_overwrite_params() handles the non-str branch so the
node and the metadata extractor conversion logic stay in sync;
unrecognized sampler functions are logged and skipped
- note: ddim is constructed by ComfyUI as euler with random inpaint,
so the ddim name is unrecoverable and extracts as euler
Downloads failed with "No suitable file found in metadata" for models whose
only file uses newer CivitAI file types (e.g. 'Enhancement LoRA' for
Anima/AIR image-editing LoRAs) because the primary-file allowlist only
covered legacy types.
- unify the weights-type allowlist as MODEL_WEIGHT_FILE_TYPES
(py/utils/constants.py) and apply it across download, recipe and
metadata-refresh lookups
- mirror CivitAI's getPrimaryFile() semantics: prefer weights-type primary,
fall back to weights files, then trust CivitAI's primary flag (excluding
non-downloadable artifacts like Config/Archive/Workflow)
- mirror the allowlist in the frontend via shared isModelWeightFile() helper
- add regression tests for the Enhancement LoRA primary-file download,
primary-flag fallback and weights-over-non-weights-primary preference
Version-tab updates reused the current version's folder, so updating a LoRA
to a version with a different base model (e.g. Illustrious -> Anima) ignored
the download path template and landed in the old version's directory.
When the target version's base model differs from the current local version
and a path template is configured, re-resolve the template under the same
model root. The backend keeps an explicitly provided root when
use_save_dir_as_root is set, so regular downloads still use the default root.
CLIP Text Encode and friends whose text widget is backed by a connected
input cannot have their text changed via the widget (execution reads the
linked input), so sending to them was a silent no-op.
- Registry: compute text_widget_connected capability from the widget's
backing input link state; has_text_widget drops to false when wired;
include the flag in the registration fingerprint so link changes
re-register the affected nodes
- Registry: hook link connect/disconnect (graph events on new litegraph,
onAfterChange fallback for classic) on root and subgraphs, plus
subgraph-created for future subgraphs
- applyWidgetUpdate: skip inject_text when the target widget is connected
and self-heal the registry instead of writing a value that is ignored
- Web UI: drop text_widget_connected nodes from prompt/embedding send
candidates; show a Mark as -> Send Prompt Target hint toast when no
candidates remain (new uiHelpers.workflow.noPromptTargets key, synced
to all locales; zh-CN/zh-TW translated)
- Extract shared resolveTextWidget() used by both the candidate-set
logic and the write path so the two cannot drift apart
- Tests: workflow registry connection-state registration, subgraph
handling, fingerprint re-registration, inject_text write/skip paths,
setup link-change hooks; uiHelpers candidate filtering and hint toast
Move the recipe delete modal's undo warning into modals.deleteRecipe.
recoverableWarning instead of hardcoded English; sync placeholders into
all 9 non-English locales
- Remove delete_undo_enabled setting (backend default, frontend state,
settings modal UI, 10 locales); staged deletes with 30s undo are now
the only delete path and stale settings keys are silently ignored
- Remove the 1500ms delete-button arm delay (armDeleteButton) from all
delete modals; misclicks are recoverable via the undo toast
- Delete modal always shows the recoverable warning
- Log the first staged file path in staging log lines for easier support
Refresh after manual .metadata.json deletion rebuilds the payload without
file_name/size/modified, which are required by BaseModelMetadata.from_dict.
The recreated sidecar then fails to parse and the scanner skips the model.
- load_metadata_payload fills missing file facts from os.stat
- hydrate_model_data restores every missing key from the cache snapshot
only when the sidecar is missing entirely (disk stays authoritative
otherwise), preferring the cached import timestamp for modified
- save_metadata fills file facts on write so no write path can produce
an unparseable sidecar
Promote aria2 stderr lines that indicate disk write failures (e.g. the
'cause: No space left on device' line following 'Write disk cache flush
failure') from DEBUG to INFO so the root cause is visible in default logs,
including Windows-specific phrases (file locked by another process, sharing
violation). The same line is rate-limited to one INFO report per 60s window
and the report map is pruned on insert so repeated failures cannot spam the
log or grow memory. All other stderr output stays at DEBUG.
Fix ~790 basedpyright errors across the test suite:
- Type stub subclasses of real production classes with super().__init__()
- Add missing generic type arguments and Dict[str, Any] annotations
- Add None guards before subscript/member access
- Adapt tests to production API changes (removed dead handlers,
PersistentModelCache.get_default, _i18n_filter_added location)
- Read AutoV3 directly from the downloaded file's own file_info hashes
(no SHA256 cross-matching against version_info.files, so the value is
captured even when the API omits SHA256)
- Extract normalize_autov3() validation helper shared with the
sha256-matching autov3_from_civitai_files path
- Fall back to the embedded safetensors header hash at download
completion; mark '' (checked-unavailable) so the startup backfill
query (autov3 IS NULL) never revisits the row
- Clear archive-level AutoV3 for zip-extracted models so per-file
header resolution applies to every extracted model
- Three-state autov3 field (not-checked / checked-unavailable / 12-hex value)
in .metadata.json sidecars, in-memory ModelHashIndex, and SQLite
(models.autov3 column + autov3_index table) with column-presence migration
- Background self-terminating backfill for legacy rows: per-model-type
concurrency guard, executor-offloaded I/O, Civitai-first resolution
(SHA256-matched version file) falling back to the embedded safetensors
header hash
- Civitai-first propagation on metadata refresh, scan, and download paths;
reject the empty-string SHA256 placeholder and strip OneTrainer 0x prefix
- List API hash filters and hash index lookups accept 12-char AutoV3
- Cap safetensors header reads at 64 MiB to prevent crafted-file allocation
- Prevent stale AutoV3 mappings on file replacement while preserving them on
same-file re-registration (lazy-hash completion)
The download modal's step shared the 'locationStep' id with the import
modal, so getElementById('locationStep') could resolve to the wrong
element depending on template include order. The import flow relied on
an injected display:block !important rule to work around it.
Rename the download modal's step id and update all references so each
modal owns a unique step id.
Make the import modal a flex column with a scrollable step area so the
Back/Import buttons stay visible on short viewports (1080p / 150% zoom)
instead of being cut off at the bottom of the scroll flow.
Also reset step scroll positions via class since 'locationStep' has a
duplicate id in the download modal template.
Add /af and /noaf toggle commands (plus /activefilters aliases) to the
loras autocomplete widget. When enabled (default off), suggestions are
matched within the active filters (folder, base model, tags, auto-tags,
license, tag logic) persisted by the LoRA Manager page in localStorage,
keeping the match pool consistent with the list endpoint, including the
global show_only_sfw setting.
Backend: /lm/{prefix}/relative-paths accepts the filter query params and
pre-filters the scanner cache with ModelFilterSet. The presence of the
recursive param signals the filter pipeline to run even without concrete
filters so global settings stay in parity with the list endpoint.
Auto-select the first (newest) version for URLs without an explicit
modelVersionId, matching the existing batch flow, so users can proceed
to location/download without manually picking a version.
os.path.commonpath raises ValueError for paths on different Windows
drives. Treat that as no common root so cross-drive recipes migrations
succeed instead of failing with 'Invalid recipes path change'.
Move the 'Download Missing' / 'Re-process All' example image actions
under a single 'Download Example Images' submenu item in the single-model
and bulk context menus, matching the existing send-to-workflow submenu
pattern. Shorten the submenu labels and update all locale translations.
Split the single-model and bulk context menu actions into 'Download
Missing Example Images' (regular endpoint, skips already-processed
models) and 'Re-process Example Images' (force endpoint, retries
failed models).
- start_download accepts model_hashes so a selected subset can be
processed with the progress-aware skip logic; explicitly targeted
models bypass the failed/processed model-level guards so per-image
gaps are filled
- pre-download existence check in the processor skips network requests
for image files already on disk across all download paths
- force download retries previously failed models and clears their
failed status on success
- add i18n keys for the new menu items across all locales
The model field now accepts either a manual string or a MODEL connection.
When wired, the model name is extracted from the patcher's
cached_patcher_init (registered by core loaders load_checkpoint_guess_config
and load_diffusion_model, preserved through LoRA clones) and converted to a
ComfyUI-style relative name via config model roots.
- model input declared as "STRING,MODEL" with widgetType STRING, so the
text widget and the dual-type connection slot coexist; non-STRING/MODEL
links are rejected by frontend and backend type validation
- UNETLoaderLM GGUF branch now registers a custom cached_patcher_init reload
factory so GGUF models participate in name extraction and ModelPatcher
deepclone/dynamic machinery
- shared collect_overwrite_params() helper keeps the node and the metadata
extractor conversion logic in sync; extraction failures are logged instead
of silently dropping the overwrite
Will Miao