musubi-tuner

Musubi Tuner

English | 日本語

Table of Contents

Click to expand

- Musubi Tuner
  - Table of Contents
  - Introduction
    - Sponsors
    - Support the Project
    - Recent Updates
    - Releases
    - For Developers Using AI Coding Agents
  - Overview
    - Hardware Requirements
    - Features
    - Documentation
  - Installation
    - pip based installation
    - Windows on ARM64
    - uv based installation
    - Linux/MacOS
    - Windows
  - Model Download
  - Usage
    - Dataset Configuration
    - Pre-caching and Training
    - Configuration of Accelerate
    - Training and Inference
  - Miscellaneous
    - SageAttention Installation
    - PyTorch version
  - Disclaimer
  - Contributing
  - License

Introduction

This repository provides scripts for training LoRA (Low-Rank Adaptation) models with HunyuanVideo, Wan2.1/2.2, FramePack, FLUX.1 Kontext, FLUX.2 dev/klein, Qwen-Image, Z-Image, and MiniMax-H3 architectures.

This repository is unofficial and not affiliated with the official repositories of these architectures.

*This repository is under development.*

Sponsors

We are grateful to the following companies for their generous sponsorship:

Support the Project

If you find this project helpful, please consider supporting its development via GitHub Sponsors. Your support is greatly appreciated!

Recent Updates

GitHub Discussions Enabled: We've enabled GitHub Discussions for community Q&A, knowledge sharing, and technical information exchange. Please use Issues for bug reports and feature requests, and Discussions for questions and sharing experiences. Join the conversation →

- September 27, 2026
    - Updated the dependencies in pyproject.toml: transformers 4.57.6 -> 5.17.0, diffusers 0.32.1 -> 0.40.0, accelerate 1.6.0 -> 1.15.0, huggingface-hub 0.34.3 -> 1.32.0. PR #1139
        - This is mainly a security maintenance update (the 4.x line of transformers and diffusers < 0.38 no longer receive fixes). The previous versions of the libraries still work with this release, so you do not have to update them immediately, but it is recommended to run pip install -e . again in your environment at your earliest convenience.
        - diffusers 0.40 requires PyTorch 2.6 or later, so PyTorch 2.6.0 or later is now required.
        - transformers 5.6 changed the internal structure of CLIPTextModel, and in 5.x CLIPTokenizer no longer applies the ftfy text normalization of the original CLIP tokenizer (straightening curly quotes, converting full-width characters, etc.). Musubi Tuner handles both, so loading CLIP-L checkpoints and the text encoder outputs of HunyuanVideo, FramePack, FLUX.1 Kontext and Kandinsky 5 are unchanged from previous versions.
        - transformers 5.6 switched the attention implementation of T5 (T5-XXL of FLUX.1 Kontext, byT5 of HunyuanVideo 1.5) to SDPA, which is faster and uses less memory. The bf16/fp16 outputs of T5 differ very slightly from previous versions (the accuracy against fp32 is the same). This may change generated images or trained weights in minor details; cached text encoder outputs from previous versions are still usable. The outputs of the other text encoders are identical.
        - The text encoder out

……（后略，见原文）