Hugging Face's logo

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Back to feed
MonsterMMORPG 
posted an update 25 days ago
Post
2587
FLUX De-Distilled and Anti-Bleeding Fine-Tuning / DreamBooth & LoRA Training Experiments

Also Testing CFG Impact for Stylized Images on base FLUX DEV model

Full size grids and full details shared here : https://www.patreon.com/posts/114969137

The aim is finding a workflow that FLUX can be trained for multiple concepts / subjects without bleeding / mixing
wikeeyang

Thanks a lot for the detail testing of the Flux Dedistilled model, I learned a lot of your test result, and I want to share my merged with Dedistilled model and finetuned version, https://huggingface.co/wikeeyang/Flux.1-Dedistilled-Mix-Tuned-fp8.

·
MonsterMMORPG

@wikeeyang awesome. and thanks a lot for comment

In this post