Based on my Centerfold Flux 5, trying to merge in Chroma and Krea.
Diffuser version 2.3 https://huggingface.co/Tiwaz/CenKreChro_V2_3
For version 2.3 I will add the quantizations under the same "tab".
Big thanks to Spooknik for making the SVDQ versions https://huggingface.co/spooknik/CenKreChro-SVDQ
Q4 KM works fine in Comfy UI but not in Forge for example.
If you need the diffuser version: https://huggingface.co/Tiwaz/CenKreChro (not 100% tested yet).
You can try using it as low as 12 steps even if I prefer 30.
Description
noting just a Q4 KM quant
FAQ
Comments (6)
I’d like to ask a question: I want to train a character LoRA. Which base model would you recommend that is compatible with this?
thanks
I did not get around to test it yet, but I did convert the model to diffusers and uploaded to hugging face, so if you use Ai toolkit you probably can train on the model. https://huggingface.co/Tiwaz/CenKreChro
but I guess there is a json missing I will check tomorrow
@solonglin342 so my hugging face is fixed, it download correctly in AI toolkit, but did not train a Lora on it yet so can't say what the result will be.
Thanks, I'll test it.
@TiwazM It works well for me. There is [a bug](https://github.com/ostris/ai-toolkit/issues/405) with AI-Toolkit that it cannot unload the TE, otherwise it was a fast and smooth process on an A40.