GitHub 热门项目：fluxgym

GitHub项目：fluxgym 仓库地址：https://github.com/cocktailpeanut/fluxgym Stars：3220 | 作者：cocktailpeanut 项目描述：Dead simple FLUX LoRA training UI with LOW VRAM support ================================================== README 内容： # Flux Gym Dead simple web UI for training FLUX LoRA **with LOW VRAM (12GB/16GB/20GB) support.** - **Frontend:** The WebUI forked from [AI-Toolkit](https://github.com/ostris/ai-toolkit) (Gradio UI created by https://x.com/multimodalart) - **Backend:** The Training script powered by [Kohya Scripts](https://github.com/kohya-ss/sd-scripts) FluxGym supports 100% of Kohya sd-scripts features through an [Advanced](#advanced) tab, which is hidden by default.  --- # What is this? 1. I wanted a super simple UI for training Flux LoRAs 2. The [AI-Toolkit](https://github.com/ostris/ai-toolkit) project is great, and the gradio UI contribution by [@multimodalart](https://x.com/multimodalart) is perfect, but the project only works for 24GB VRAM. 3. [Kohya Scripts](https://github.com/kohya-ss/sd-scripts) are very flexible and powerful for training FLUX, but you need to run in terminal. 4. What if you could have the simplicity of AI-Toolkit WebUI and the flexibility of Kohya Scripts? 5. Flux Gym was born. Supports 12GB, 16GB, 20GB VRAMs, and extensible since it uses Kohya Scripts underneath. --- # News - 9月25日：Docker支持+自动下载模型（设置时无需手动下载模型）+支持自定义基础模型（不仅仅是flux-dev，而是任何东西，只需要包含在[models.yaml](models.yaml)文件中。 - 9 月 16 日：添加“发布到 Huggingface”+ 100% Kohya sd-scripts 功能支持：https://x.com/cocktailpeanut/status/1835719701172756592 - 9 月 11 日：自动生成示例图像 + 自定义分辨率：https://x.com/cocktailpeanut/status/1833881392482066638 --- # 支持的型号 1. Flux1-dev 2. Flux1-dev2pro（如下所述：https://medium.com/@zhiwangshi28/why-flux-lora-so-hard-to-train-and-how-to-overcome-it-a0c70bc59eaf） 3. Flux1-schnell（无法获得高质量结果，因此不推荐，但请随意尝试） 4.更多？ 当您使用所选模型开始训练时，会自动下载模型。 您可以通过编辑 [models.yaml](models.yaml) 文件轻松地将更多内容添加到支持的模型列表中。如果您想分享一些有趣的基础模型，请发送 PR。 --- # 人们如何使用 Fluxgym 以下是使用 Fluxgym 本地训练 Lora 的人分享他们的经验： https://pinokio.computer/item?uri=https://github.com/cocktailpeanut/fluxgym # 更多信息 要了解更多信息，请查看此 X 线程：https://x.com/cocktailpeanut/status/1832084951115972653 # 安装 ## 1. 一键安装 您可以使用 Pinokio 一键启动器在本地自动安装和启动所有内容：https://pinokio.computer/item?uri=https://github.com/cocktailpeanut/fluxgym ## 2. 手动安装 首先克隆 Fluxgym 和 kohya-ss/sd-scripts： ```` git 克隆 https://github.com/cocktailpeanut/fluxgym cd Fluxgym git克隆-b sd3 https://github.com/kohya-ss/sd-scripts ```` 您的文件夹结构将如下所示： ```` /fluxgym 应用程序.py 要求