Chroma Modular WF v1.2
Chroma model, by Lodestone ( https://huggingface.co/lodestones/Chroma ) is getting more and more interesting day after day... and it is already getting a lot of interest.
Today (June 10th), we are at version v35, but as you may already know, it's still training. So we have just a "beta-version" model.It is based on Flux Schnell, but it is uncensored, it is lighter, completely open-source, and can use even negative prompts.
It is already available on quantized model version:
- FP8 scaled: https://huggingface.co/Clybius/Chroma-fp8-scaled
- GGUF quantized: https://huggingface.co/silveroxides/Chroma-GGUF
The workflow allows you to do many things: txt2img or img2img, inpaint (with limitation), HiRes Fix, FaceDetailer, Ultimate SD Upscale, Postprocessing and Save Image with Metadata.
The User Interface was completely re-designed, trying to keep the needed custom nodes to the bare minimum (I used only the most common and used ComfyUI custom nodes available).
Hope you will enjoy the wf and the Chroma model, as it is really promising... Trust me, keep an eye on it!
-------------------------------
Chroma: Open-Source, Uncensored, and Built for the Community
Chroma is a 8.9B parameter model based on FLUX.1-schnell (technical report coming soon!). It’s fully Apache 2.0 licensed, ensuring that anyone can use, modify, and build on top of it—no corporate gatekeeping.
The model is still training right now, and I’d love to hear your thoughts! Your input and feedback are really appreciated.
What Chroma Aims to Do
- Training on a 5M dataset, curated from 20M samples including anime, furry, artistic stuff, and photos.
- Fully uncensored, reintroducing missing anatomical concepts.
- Built as a reliable open-source option for those who need it.