Chroma: Open-Source, Uncensored, and Built for the Community - [WIP]
Hey everyone!
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.
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.
See the Progress
Hugging Face Repo: https://huggingface.co/lodestones/Chroma
Hugging Face Debug Repo: https://huggingface.co/lodestones/chroma-debug-development-only
Live WandB Training Logs: https://wandb.ai/lodestone-rock/optimal%20transport%20unlocked
ComfyUI Inference node [WIP]: https://github.com/lodestone-rock/flux-mod
Training code!: https://github.com/lodestone-rock/flow
Support Open-Source AI
The current pretraining run has already used 5000+ H100 hours, and keeping this going long-term is expensive.
If you believe in accessible, community-driven AI, any support would be greatly appreciated.
👉 [https://ko-fi.com/lodestonerock] — Every bit helps!
this looks very exciting! large flux finetunes are so rare.
Been using this for a small bit, very impressive and diverse!
Remember to use the FluxMod ComfyUI nodes to load the model & utilize padding removal on the T5 conditioning for proper inference!
How does padding removal work? I tried applying it to the conditioning outputs of my positive and negative prompt and only got an error that said 'attention_mask'.
Nvm, I got it.
I got the same error. I can't use it. Is there a way to make it work?
Yes.
