🚀 Run WAN 2.x with ComfyUI and provisioning — RunPod

A streamlined and automated environment for running ComfyUI with WAN 2.x video models, optimized for use on RunPod.

• Automatic model and LoRA downloads via environment variables.
• Built-in authentication for:
  • ComfyUI
  • Code Server
  • Hugging Face API
  • CivitAI API
• Supports advanced workflows for video generation and enhancement using pre-installed custom nodes.
• Compatible with high-performance NVIDIA GPUs.

• All available templates on RunPod are tested on an L40S/A40.

👉 One-click Deploy on RunPod WAN 2.2 t2v (lightx2v)

👉 One-click Deploy on RunPod WAN 2.2 i2v + SVI 2.0 PRO (lightx2v)

👉 One-click Deploy on RunPod WAN 2.2 animate (lightx2v)

👉 One-click Deploy on RunPod WAN SCAIL-2 VI2V (lightx2v)

• ⚙️ Start
• 📚 Tutorial
• ⚙️ Provisioning examples

• PyTorch Runtime

• ComfyUI Runtime

docker pull ls250824/run-comfyui-wan2:<tag>

Use a local Docker setup to build the image using the included Python script.

git clone https://github.com/jalberty2018/run-comfyui-wan2.git
cp ./run-comfyui-wan2/build_docker.py ..

export DOCKER_BUILDKIT=1
export COMPOSE_DOCKER_CLI_BUILD=1

python3 build_docker.py --username=<your_dockerhub_username> --tag=<custom_tag> --latest run-comfyui-wan2

Original code, scripts, configuration and documentation in this repository are licensed under the MIT License, except where otherwise noted. Third-party software and model weights retain their own licenses; MIT does not apply to the complete assembled Docker image. See Third-party notices.

Model weights are downloaded separately, including automatically through the user's provisioning settings. Users must review model licenses and applicable restrictions before deployment and are responsible for their inputs, use and sharing of generated content. See Model usage and responsibility.