PixFuz β€” Pixel Art Generator Hub

One Gradio app that catalogs and runs every pixel-art text-to-image model found on the Hugging Face Hub (LoRAs + checkpoints), via Inference Providers.

The bundled models_catalog.json currently lists 300+ Hub repos discovered with exhaustive search (pixel art, pixelart, spritesheet / FLUX / SDXL / Z-Image / Qwen / Krea variants, and follow-up saturation passes).

Run locally

pip install -r requirements.txt
set HF_TOKEN=hf_your_token
python app.py

Generation uses Inference Providers (default fal-ai). You need provider credits or a PRO plan with included credits: https://huggingface.co/settings/billing

Deploy as a Space

Hosting a Gradio Space on free cpu-basic currently requires a Hugging Face PRO subscription. After upgrading:

hf repos create YOUR_USER/PixFuz --type space --space-sdk gradio --public
hf spaces secrets add YOUR_USER/PixFuz --secrets HF_TOKEN
hf upload YOUR_USER/PixFuz . --type space

Refresh the model list

set HF_TOKEN=hf_...
python discover_models.py
python finalize_catalog.py
python add_missing.py
python add_wave2.py
python saturate_search.py
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support