https://github.com/techcow2/wan2gp-one-click-script
A one-click, fully automated script for installing Wan2GP.
https://github.com/techcow2/wan2gp-one-click-script
artificial-intelligence batch-script lowvram nvidia-gpu oneclick texttovideo wan21
Last synced: 10 months ago
JSON representation
A one-click, fully automated script for installing Wan2GP.
- Host: GitHub
- URL: https://github.com/techcow2/wan2gp-one-click-script
- Owner: techcow2
- License: mit
- Created: 2025-06-02T12:05:59.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-06-02T14:56:02.000Z (about 1 year ago)
- Last Synced: 2025-06-27T00:50:19.347Z (12 months ago)
- Topics: artificial-intelligence, batch-script, lowvram, nvidia-gpu, oneclick, texttovideo, wan21
- Language: Batchfile
- Homepage:
- Size: 28.3 KB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Wan2GP One-Click Automated Installer
This repository provides a one-click, fully automated script for installing [Wan2GP](https://github.com/deepbeepmeep/Wan2GP), an open-source text-to-video and Lora generation suite for NVIDIA GPUs. **Wan2GP is built for the GPU poor—offering low VRAM requirements, fast generation, and an easy-to-use web UI.** The script manages all dependencies, sets up the environment, and launches Wan2GP in your browser.
---
## Features
- Automated installation and launch process
- Detection and installation of Python 3.10 and Git if not present
- NVIDIA GPU detection and selection of compatible CUDA/PyTorch versions
- Setup of all required and optional dependencies (including SageAttention, Triton, and Flash Attention)
- Automatic fallback to compatible attention methods if needed
- No manual setup required
---
## Quick Start
1. Download the `Wan2GP_OneClick_Installer.bat` file from this repository.
2. Right-click the file and select **Run as Administrator**.
3. The script will:
- Check for Python 3.10 and Git, installing them if needed
- Detect your NVIDIA GPU and select the appropriate PyTorch and CUDA version
- Clone the Wan2GP repository
- Set up a Python virtual environment and install all dependencies
- Attempt to optimize with SageAttention, Triton, and Flash Attention (if supported)
- Launch Wan2GP in your browser at [http://localhost:7860](http://localhost:7860)
To relaunch Wan2GP later, run the installer again. It will skip completed installation steps if already set up.
---
## Requirements
- Windows 10/11 (64-bit)
- NVIDIA GPU with CUDA support (GTX 600 series or newer recommended)
- Internet connection
---
## Script Workflow
| Step | Action |
|-----------------------|------------------------------------------------------------------------|
| Python & Git Check | Installs Python 3.10.9 and Git if not present |
| GPU Detection | Detects NVIDIA GPU and selects compatible PyTorch/CUDA version |
| Source Download | Clones the Wan2GP repository |
| Environment Setup | Creates and activates a Python virtual environment |
| Dependency Install | Installs PyTorch, Python requirements, and performance enhancements |
| Launch | Opens Wan2GP in your browser with the selected attention method |
If a step fails (such as advanced attention methods), the script will automatically use a compatible alternative.
---
## Troubleshooting
- If no NVIDIA GPU is detected, the script will notify you. Wan2GP requires an NVIDIA GPU for full functionality.
- Ensure the latest NVIDIA drivers are installed.
- A stable internet connection is required for downloading dependencies.
---
## Customization
- The script is designed for automated use, but advanced users can modify it to change Python or CUDA versions, or add custom dependencies.
- All logic is contained in the batch script.
---
## About Wan2GP
Wan2GP is an open-source text-to-video and Lora generation toolkit featuring:
- Video generation with low VRAM requirements
- Web-based user interface
- Support for custom and prebuilt Lora packs
- Optimized for NVIDIA GPUs
See the [Wan2GP GitHub repository](https://github.com/deepbeepmeep/Wan2GP) for more information.
---
## License
This installer script is released under the [MIT License](LICENSE). You are free to use, modify, and distribute this script with proper attribution.