mirror of https://github.com/willmiao/ComfyUI-Lora-Manager.git synced 2026-07-05 17:01:16 -03:00

Files

Will Miao 5494a70f40 chore(tests): commit validation dataset and baseline reports into repo

Move the HF model list from ~/Documents/ into tests/enrich_hf_validation/test_data/
and commit the pipeline validation baseline artifacts (report.json,
preprocessing_audit.json, README snapshots) into baselines/.

Update config.py and run_validation.py defaults to use repo-relative paths
via os.path.dirname(__file__) instead of ~/Documents/ hardcode.

Originates from changes in 8fb00998 (validation pipeline audit).

2026-07-05 17:03:45 +08:00

2.7 KiB

Raw Blame History

tags, widget, base_model, instance_prompt, license

tags

widget

base_model

instance_prompt

license

text-to-image

lora

diffusers

template:diffusion-lora

Retro

Pixel

text

output

Retro Pixel, A pixelated image of a german shepherd dog. The dogs fur is a vibrant shade of brown, with a black stripe running down its back. The background is a light green, and the dogs shadow is cast on the ground.

url
images/RP1.png

text

output

Retro Pixel, A pixelated image of a man surfing on a surfboard. The mans body is covered in a red shirt and blue shorts. His arms are out to the sides of his body. The surfboard is a vibrant blue color. The water is a light blue color with white splashes. The sun is shining on the right side of the image.

url
images/RP2.png

text

output

Retro Pixel, pixel art of a Hamburger in the style of an old video game, hero, pixelated 8bit, final boss

url
images/RP3.png

black-forest-labs/FLUX.1-dev

Retro Pixel

creativeml-openrail-m

Retro-Pixel-Flux-LoRA

Hosted Here🧨: https://huggingface.co/spaces/prithivMLmods/FLUX-LoRA-DLC

The model is still in the training phase. This is not the final version and may contain artifacts and perform poorly in some cases.

Model description

prithivMLmods/Retro-Pixel-Flux-LoRA

Image Processing Parameters

Parameter	Value	Parameter	Value
LR Scheduler	constant	Noise Offset	0.03
Optimizer	AdamW	Multires Noise Discount	0.1
Network Dim	64	Multires Noise Iterations	10
Network Alpha	32	Repeat & Steps	24 & 2340
Epoch	15	Save Every N Epochs	1

Labeling: florence2-en(natural language & English)

Total Images Used for Training : 16 [ Hi-RES ]

Best Dimensions

1024 x 1024 (Default)

Setting Up

import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "prithivMLmods/Retro-Pixel-Flux-LoRA"
trigger_word = "Retro Pixel"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

Trigger words

You should use Retro Pixel to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.

2.7 KiB Raw Blame History

Retro-Pixel-Flux-LoRA

Model description

Best Dimensions

Setting Up

Trigger words

Download model

2.7 KiB

Raw Blame History