Wan MCP
Generate videos from text or images using Wan in the AI Agent through MCP.
Turn text and reference assets into videos, with first/last frame control and multimedia references. Integrate into your own creative workflow through a unified API.
One API covers text-to-video, first/last frames, reference images, reference videos, and audio-driven generation, automatically selecting the appropriate generation mode.
Generate complete shots from prompts, ideal for concept videos, social content, and video drafts.
Lock the opening and closing frames for greater control over motion, subject positioning, and transition direction.
Use images, videos, or audio to reference characters, actions, styles, and rhythm without switching APIs.
Both videos were generated by the production API and transferred to the AceDataCloud CDN for direct playback and verification.
curl -X POST 'https://api.acedata.cloud/wan/videos' \
-H 'Authorization: Bearer YOUR_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"model": "wan3.0-video",
"prompt": "A red ball on a white studio floor",
"duration": 2,
"resolution": "480P",
"ratio": "16:9",
"async": true
}'{
"success": true,
"task_id": "9478f7ee-5ebb-47ac-ab49-63fbb42d8233",
"data": {
"status": "succeeded",
"usage": {
"input_video_duration": 0,
"output_video_duration": 2,
"SR": 480,
"ratio": "16:9"
}
},
"cost": {"list_amount": 1.16}
}Production verification trace: 087bbe73-e336-4b4a-a410-c56d54c763b8。Async tasks can be queried through theTasks API, and polling does not incur duplicate charges.
The Pricing tab displays the standard price per second by resolution; the final bill is calculated based on the actual input video seconds and successful output seconds in the response usage.
Ideal for quick previews, shot drafts, and mobile content.
Ideal for everyday publishing, social content, and product demonstrations.
Ideal for high-quality final videos, detailed visuals, and large-screen displays.
Credit and USD prices per second are displayed in real time on this page's Pricing tab; static prices are not duplicated in the introductory text.
Enables smart duration. Pre-authorization is checked against an upper limit of 30 seconds, and final billing is still based on actual usage after completion.
Input reference video seconds and successful output video seconds are added together; image, audio, and text inputs do not add video seconds.
See the Wan Video Generation API documentation; field constraints and response structures are governed by this documentation.
Start with text-based shots, then add frame, video, image, and audio references as needed.
Get API Key