kohya-ss / kohya-ss/sd-scripts

any plans to support training for OpenFlux?

Open
#1,766 3 comments 2 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
7.2k
Forks
1.2k
Avg merge
11m
Merged PRs (30d)
2

Description

@kohya-ss @lansing @rockerBOO @akx @tsukimiya @wkpark
Would you consider supporting training for OpenFlux? The OpenFlux model link is: https://huggingface.co/ostris/OpenFLUX.1. Given that Flux and its derivatives have strict commercial usage restrictions—including limitations on combining SD3.0 with SD3.5 for commercial purposes—OpenFlux stands out as an option. Currently, only the Schnell version of Flux is available for commercial use, but its performance is relatively poor, and its guidance distillation and step distillation make fine-tuning challenging. Currently, OpenFlux is the only open-source model with decent performance, based on the de-distilled version of Schnell, and it has attracted significant attention. Do you have any plans to support training for OpenFlux?

Contributor guide

No contributing guide indexed for this repository

Research direction

No files, tests, or entry points are named. Start by reviewing the repository's existing Flux training support and the OpenFLUX.1 model at the linked Hugging Face page. Done would require an agreed implementation scope for training OpenFlux, but this issue does not define acceptance criteria.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.