cocktailpeanut/fluxgym
A Gradio-based web UI that wraps Kohya Scripts to enable low-VRAM FLUX LoRA fine-tuning.
Not currently ranked — collecting fresh signals.
star history
FluxGym provides a graphical interface for training Low-Rank Adaptation (LoRA) adapters on FLUX image-generation diffusion models. It wraps Kohya sd-scripts training logic behind a Gradio frontend forked from AI-Toolkit, enabling training on consumer GPUs with as little as 12GB VRAM. Users can configure training parameters, generate sample images during training, and publish results directly to Hugging Face.
Frequently asked
- What is cocktailpeanut/fluxgym?
- A Gradio-based web UI that wraps Kohya Scripts to enable low-VRAM FLUX LoRA fine-tuning.
- Is fluxgym open source?
- Yes — cocktailpeanut/fluxgym is open source, released under the MIT license.
- What language is fluxgym written in?
- cocktailpeanut/fluxgym is primarily written in Python.
- How popular is fluxgym?
- cocktailpeanut/fluxgym has 3.2k stars on GitHub.
- Where can I find fluxgym?
- cocktailpeanut/fluxgym is on GitHub at https://github.com/cocktailpeanut/fluxgym.