OSS Alternative - Discover Top Open Source Alternatives to Popular Software

Nerogar/OneTrainer

A comprehensive, one-stop solution for training various diffusion models with advanced features and a user-friendly interface.

Core Features

Supports a wide range of diffusion models (SD, SDXL, Flux, etc.) and formats.

Offers multiple training methods (LoRA, fine-tuning) with advanced features like masked training and aspect ratio bucketing.

Integrates dataset and model tooling for captioning, mask creation, model conversion, and in-training sampling.

Provides dual GUI and CLI modes, complemented by automatic backups and TensorBoard integration.

Quick Start

git clone https://github.com/Nerogar/OneTrainer.git

Detailed Introduction

OneTrainer is an open-source, comprehensive platform designed to streamline the training process for a wide array of diffusion models. It offers a unified environment for fine-tuning models like Stable Diffusion, SDXL, and Flux, supporting various training methods including LoRA and full fine-tuning. Beyond core training, it provides essential tooling for dataset preparation (captioning, masking) and model management (conversion), alongside features like automatic backups and TensorBoard integration. Its dual GUI and CLI modes cater to both beginners and advanced users, making complex AI model training accessible and efficient.