Wan 3.0 ComfyUI

ComfyUI custom nodes for the Wan 3.0 API, delivered through Muapi.

The pack follows the focused Muapi workflow pattern: add one API Key node, connect it to a Wan 3.0 generation node, then preview the returned IMAGE output or save the returned media URL into ComfyUI/output.

▶ Watch: How to Access Wan 3.0 API - Best Uncensored Alternative to Seedance 2

Related Projects

• Wan-3.0-API — Python SDK and MCP server for Wan 3.0-compatible video workflows.
• Wan-3.0-Prime-API — Python SDK and MCP server for the higher-fidelity Wan 3.0 Prime tier.
• Open-Generative-AI — Open-source media studio that uses MuAPI for image and video workflows.
• muapi-comfyui — General MuAPI ComfyUI node pack.
• seedance2-comfyui — Seedance 2 custom nodes and workflows for ComfyUI.
• seedance2.5-comfyui — Native Seedance 2.5 custom nodes and workflows for ComfyUI.
• veo3.1-comfyui — Veo 3.1 custom nodes and workflows for ComfyUI.
• FLUX 3 ComfyUI — focused Muapi nodes for FLUX 3 image and video workflows.
• MiniMax H3 ComfyUI — focused Muapi nodes for MiniMax H3 video workflows.
• Muapi API reference — shared authentication, polling, and upload behavior.

Included nodes

Node	What it does
Wan 3.0 API Key	Stores one Muapi API key for the workflow.
Wan 3.0 Text-to-Video	Generates a video with synchronized audio from a text prompt (resolution, aspect ratio, duration, thinking mode, seed).
Wan 3.0 Image-to-Video	Animates one source image with a motion prompt, with optional end-frame guidance and the same audio/thinking-mode/duration controls.
Wan 3.0 Reference-to-Video	Generates a video guided by up to 10 reference images, 5 reference videos, and 5 reference audios.
Wan 3.0 Text-to-Image	Generates an image from a text prompt. Coming soon — registered but not yet live on Muapi.
Wan 3.0 Image Edit	Applies an instruction prompt to a source image. Coming soon — registered but not yet live on Muapi.
Wan 3.0 Save Image	Downloads an image URL to ComfyUI/output and returns an IMAGE.
Wan 3.0 Save Video	Downloads a video URL to ComfyUI/output and returns decoded frames.

Installation

Copy this directory into ComfyUI/custom_nodes/wan-3.0-comfyui, then install the dependencies in the same Python environment ComfyUI uses:

pip install -r ComfyUI/custom_nodes/wan-3.0-comfyui/requirements.txt

Restart ComfyUI. The example workflows can be loaded with File → Load.

Included examples:

• Wan30_T2I_Example.json
• Wan30_I2I_Example.json
• Wan30_T2V_Example.json
• Wan30_I2V_Example.json

API key

Use the 🔑 Wan 3.0 API Key node and connect its output to a generation node. Alternatively, leave the node field empty and configure one of these:

export MUAPI_API_KEY="your_muapi_key"

or:

muapi auth configure --api-key YOUR_KEY

The nodes also read ~/.muapi/config.json when it contains an api_key field. Keys are only used for requests and are not written into generated workflow files unless you explicitly save them in a node field.

Quick workflows

Text-to-video

1. Add 🔑 Wan 3.0 API Key.
2. Add 🎬 Wan 3.0 Text-to-Video. Set resolution, aspect_ratio, duration (2-30s), thinking_mode, enable_audio, and seed as needed.
3. Connect video_url to 🎬 Wan 3.0 Save Video and first_frame to Preview Image.

Image-to-video

Connect Load Image → IMAGE to 🎬 Wan 3.0 Image-to-Video, or provide a hosted URL/local path in image_url. The node uploads a connected ComfyUI image before submitting the generation request. Use last_image_url to steer the generated video toward a specific end frame.

Reference-to-video

Add 🎬 Wan 3.0 Reference-to-Video and list one or more reference URLs (newline- or comma-separated) in reference_images (up to 10), reference_videos (up to 5), and/or reference_audios (up to 5). Reference media are identified by their order within each field, so the prompt can refer to "the first reference image" or similar.

Text-to-image / Image edit (coming soon)

🖼️ Wan 3.0 Text-to-Image and 🖌️ Wan 3.0 Image Edit are included for when Muapi ships these endpoints, but calls to them currently fail — the underlying Wan 3.0 image models are not yet live.

API contract

Workflow	Endpoint	Request body
Text-to-video	POST /api/v1/wan3.0-text-to-video	prompt, resolution, aspect_ratio, duration, thinking_mode, enable_audio, seed
Image-to-video	POST /api/v1/wan3.0-image-to-video	prompt, image_url, last_image (optional), plus the same video params
Reference-to-video	POST /api/v1/wan3.0-reference-to-video	prompt, images_list (≤10), videos_list (≤5), audios_list (≤5), plus the same video params
Text-to-image (coming soon)	POST /api/v1/wan3.0-text-to-image	prompt
Image edit (coming soon)	POST /api/v1/wan3.0-image-edit	prompt, image_url
Poll task	GET /api/v1/predictions/{request_id}/result	—
Upload local image	POST /api/v1/upload_file	multipart file

All requests use the x-api-key header. Generation is asynchronous: the nodes submit a task, poll until it completes, and return the final URL plus an IMAGE tensor (the generated image or the video's first frame).

For local testing against a compatible Muapi deployment, set WAN_3_API_BASE_URL. Polling can be tuned with WAN_3_POLL_INTERVAL and WAN_3_MAX_WAIT.

Requirements

Python 3.8+, requests, Pillow, numpy, and opencv-python. ComfyUI supplies the installed PyTorch runtime used by IMAGE tensors. For in-ComfyUI video playback, install ComfyUI-VideoHelperSuite.

License

MIT