{"name":"Fluxgym: Simple FLUX LoRA Training UI with Low VRAM Support","description":"Fluxgym offers a user-friendly web interface for training FLUX LoRA models, specifically designed to support systems with low VRAM, such as 12GB, 16GB, and 20GB GPUs. It combines the simplicity of a Gradio UI, forked from AI-Toolkit, with the powerful and flexible training capabilities of Kohya sd-scripts. This tool allows users to easily train custom LoRAs, including advanced features like automatic sample image generation and direct publishing to Hugging Face.","github":"https://github.com/cocktailpeanut/fluxgym","url":"https://osrepos.com/repo/cocktailpeanut-fluxgym","source":"osrepos.com","sourceDescription":"This repository profile is provided by osrepos.com, an open source repository discovery platform.","repositoryProfile":"https://osrepos.com/repo/cocktailpeanut-fluxgym","generatedFor":"open source discovery and AI-assisted research","markdown":"https://osrepos.com/repo/cocktailpeanut-fluxgym.md","json":"https://osrepos.com/repo/cocktailpeanut-fluxgym.json","topics":["Python","AI","Machine Learning","Generative AI","LoRA Training","FLUX","Stable Diffusion","Low VRAM"],"keywords":["Python","AI","Machine Learning","Generative AI","LoRA Training","FLUX","Stable Diffusion","Low VRAM"],"stars":null,"summary":"Fluxgym offers a user-friendly web interface for training FLUX LoRA models, specifically designed to support systems with low VRAM, such as 12GB, 16GB, and 20GB GPUs. It combines the simplicity of a Gradio UI, forked from AI-Toolkit, with the powerful and flexible training capabilities of Kohya sd-scripts. This tool allows users to easily train custom LoRAs, including advanced features like automatic sample image generation and direct publishing to Hugging Face.","content":"## Introduction\nFluxgym is an intuitive web user interface designed to simplify the process of training FLUX LoRA models, especially for users with limited VRAM. It provides robust support for GPUs with 12GB, 16GB, and 20GB of VRAM, making advanced model training more accessible. The project integrates a user-friendly Gradio frontend, inspired by AI-Toolkit, with the highly flexible and powerful backend of Kohya sd-scripts, ensuring full compatibility with all Kohya features.\n\n## Installation\nFluxgym offers multiple convenient installation methods:\n\n### One-Click Install\nFor the easiest setup, you can use the Pinokio 1-click launcher:\n[https://pinokio.computer/item?uri=https://github.com/cocktailpeanut/fluxgym](https://pinokio.computer/item?uri=https://github.com/cocktailpeanut/fluxgym){:target=\"_blank\"}\n\n### Manual Installation\n1.  **Clone Repositories:**\n    bash\ngit clone https://github.com/cocktailpeanut/fluxgym\ncd fluxgym\ngit clone -b sd3 https://github.com/kohya-ss/sd-scripts\n    \n2.  **Create and Activate Virtual Environment:**\n    *   Windows:\n        bash\npython -m venv env\nenv\\Scripts\\activate\n        \n    *   Linux:\n        bash\npython -m venv env\nsource env/bin/activate\n        \n3.  **Install Dependencies:**\n    bash\ncd sd-scripts\npip install -r requirements.txt\ncd ..\npip install -r requirements.txt\n    \n4.  **Install PyTorch Nightly:**\n    *   For most GPUs (cu121):\n        bash\npip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121\n        \n    *   For NVIDIA RTX 50-series (cu128):\n        bash\npip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu128\npip install -U bitsandbytes\n        \n5.  **Start Fluxgym:**\n    Ensure your virtual environment is activated, then run:\n    bash\npython app.py\n    \n\n### Docker Installation\n1.  **Clone Repositories:**\n    bash\ngit clone https://github.com/cocktailpeanut/fluxgym\ncd fluxgym\ngit clone -b sd3 https://github.com/kohya-ss/sd-scripts\n    \n2.  **Build and Run with Docker Compose:**\n    Adjust `PUID` and `PGID` environment variables if your user ID and group ID are not 1000.\n    bash\ndocker compose up -d --build\n    \n    Access the UI at [http://localhost:7860](http://localhost:7860){:target=\"_blank\"} in your web browser.\n\n## Examples\nUsing Fluxgym is designed to be straightforward:\n1.  Enter your LoRA information.\n2.  Upload your training images and caption them, including your chosen trigger word.\n3.  Click \"start\" to begin training.\n\nFluxgym also includes powerful configuration options:\n*   **Automatic Sample Image Generation:** Configure prompts and frequency to generate sample images during training, allowing you to monitor the LoRA's evolution. Advanced flags like `--d` (seed) can be used for consistent comparisons.\n*   **Publishing to Hugging Face:** Easily log in with your Hugging Face token and publish your trained LoRA models directly from the UI.\n*   **Advanced Kohya Features:** An \"Advanced\" tab, hidden by default, exposes all underlying Kohya sd-scripts launch flags, providing full control over the training process.\n*   **Caption File Uploads:** Supports uploading `.txt` caption files alongside corresponding image files.\n\n## Why Use Fluxgym?\nFluxgym addresses several key challenges in FLUX LoRA training:\n*   **Low VRAM Accessibility:** It democratizes FLUX LoRA training by supporting GPUs with as little as 12GB VRAM, which is a significant barrier for many users.\n*   **Simplified User Experience:** It provides a \"dead simple\" web UI, abstracting away the complexities of command-line arguments, making training accessible to a wider audience.\n*   **Full Kohya Script Power:** By leveraging Kohya sd-scripts as its backend, Fluxgym ensures users have access to all advanced features and flexibility, without sacrificing ease of use.\n*   **Streamlined Workflow:** Features like automatic model downloads, sample image generation, and direct Hugging Face publishing streamline the entire LoRA training and sharing process.\n\n## Links\n*   **GitHub Repository:** [https://github.com/cocktailpeanut/fluxgym](https://github.com/cocktailpeanut/fluxgym){:target=\"_blank\"}\n*   **Pinokio One-Click Install:** [https://pinokio.computer/item?uri=https://github.com/cocktailpeanut/fluxgym](https://pinokio.computer/item?uri=https://github.com/cocktailpeanut/fluxgym){:target=\"_blank\"}\n*   **More Info (X Thread):** [https://x.com/cocktailpeanut/status/1832084951115972653](https://x.com/cocktailpeanut/status/1832084951115972653){:target=\"_blank\"}","metrics":{"detailViews":7,"githubClicks":2},"dates":{"published":null,"modified":"2025-10-12T20:42:38.000Z"}}