• SD-Trainer
• • ✨NEW: Train WebUI
• Usage
• • • Required Dependencies
    • Clone repo with submodules
  • ✨ SD-Trainer GUI
  • • Windows
    • • Installation
      • Train
    • Linux
    • • Installation
      • Train
  • Legacy training through run script manually
  • • Windows
    • • Installation
      • Train
    • Linux
    • • Installation
      • Train
      • TensorBoard
  • Program arguments

SD-Trainer

✨ Enjoy Stable Diffusion Train！ ✨

Download · Documents · 中文README

LoRA-scripts (a.k.a SD-Trainer)

LoRA & Dreambooth training GUI & scripts preset & one key training environment for kohya-ss/sd-scripts

✨NEW: Train WebUI

The REAL Stable Diffusion Training Studio. Everything in one WebUI.

Follow the installation guide below to install the GUI, then run run_gui.ps1(windows) or run_gui.sh(linux) to start the GUI.

Tensorboard	WD 1.4 Tagger	Tag Editor

Usage

Required Dependencies

Python 3.10 and Git

Clone repo with submodules

git clone --recurse-submodules https://github.com/Akegarasu/lora-scripts

✨ SD-Trainer GUI

Windows

Installation

Run install.ps1 will automaticilly create a venv for you and install necessary deps.
If you are in China mainland, please use install-cn.ps1

Train

run run_gui.ps1, then program will open http://127.0.0.1:28000 automanticlly

Linux

Installation

Run install.bash will create a venv and install necessary deps.

Train

run bash run_gui.bash, then program will open http://127.0.0.1:28000 automanticlly

Legacy training through run script manually

Windows

Installation

Run install.ps1 will automaticilly create a venv for you and install necessary deps.

Train

Edit train.ps1, and run it.

Linux

Installation

Run install.bash will create a venv and install necessary deps.

Train

Training script train.sh will not activate venv for you. You should activate venv first.

source venv/bin/activate

Edit train.sh, and run it.

TensorBoard

Run tensorboard.ps1 will start TensorBoard at http://localhost:6006/

Program arguments

Parameter Name	Type	Default Value	Description
--host	str	"127.0.0.1"	Hostname for the server
--port	int	28000	Port to run the server
--listen	bool	false	Enable listening mode for the server
--skip-prepare-environment	bool	false	Skip the environment preparation step
--disable-tensorboard	bool	false	Disable TensorBoard
--disable-tageditor	bool	false	Disable tag editor
--tensorboard-host	str	"127.0.0.1"	Host to run TensorBoard
--tensorboard-port	int	6006	Port to run TensorBoard
--localization	str		Localization settings for the interface
--dev	bool	false	Developer mode to disale some checks