ComfyUI-VeniceWAN

ComfyUI-VeniceWAN

Generate videos with Wan 3.0 through Venice.ai, directly in ComfyUI.

• Text-to-Video — generate from a prompt.
• Image-to-Video — animate a starting image, with an optional last frame.
• Reference-to-Video — use reference images, videos, and audio.

Models and their supported resolution, aspect-ratio, duration, and audio controls load from Venice. All three nodes return native ComfyUI VIDEO output. Requires a Venice API key and credits.

Install

In ComfyUI Manager → Install via Git URL, enter:

https://github.com/nakedfighter3d/ComfyUI-VeniceWAN

Or run from your ComfyUI directory:

git clone https://github.com/nakedfighter3d/ComfyUI-VeniceWAN.git custom_nodes/ComfyUI-VeniceWAN
python -m pip install -r custom_nodes/ComfyUI-VeniceWAN/requirements.txt

Use ComfyUI's Python for the installation. Restart ComfyUI after installing or updating.

Set up and use

1. Open Settings → VeniceWAN.
2. Enter your Venice API key and click Save & connect.
3. Add a VeniceWAN node, select a model, and connect its output to Save Video.

The key is saved locally on the ComfyUI machine and kept out of workflows. You can also use VENICE_API_KEY instead; an environment key takes precedence over the Settings field.

Use a current ComfyUI release with native VIDEO support. Last-frame conditioning depends on the selected Venice model; live support across all Wan 3 variants has not been verified.