Wan 2.2 Animate Move API

wan-video/wan-animate/move

Animate a still character image with the motion and facial expressions of a driving video — no text prompt needed, just a character photo and a reference clip.

0.15 per second of output video

OpenAPI

Input

Video URL

URL of the driving video. Its body motion and facial expressions are transferred onto the character image.

Image URL

URL of the character image to animate. It is resized and center-cropped to the driving video's aspect ratio.

Resolution

Output resolution. 480p (default) is the cheapest tier, 580p mid, 720p the highest quality.

Additional Settings

Customize your input with more control.

Guidance Scale

Min: 1 - Max: 10

Classifier-free guidance strength. Higher values follow the conditioning more strictly; the default of 1 suits most driving videos.

Num Inference Steps

Min: 2 - Max: 40

Number of denoising steps. More steps can add detail at the cost of runtime.

Shift

Min: 1 - Max: 10

Flow-matching timestep shift. Leave at the default unless tuning motion fidelity.

Video Quality

Encoding quality of the returned mp4.

Video Write Mode

Trade-off used when writing the mp4: 'fast' writes quickest, 'small' produces the smallest file, 'balanced' sits between them.

Return Frames ZIP

Whether to also produce a ZIP archive of the generated frames.

Use Turbo

Run the accelerated variant: faster generation with a slight reduction in fidelity.

Seed

Random seed. Leave unset for a random result; reuse a value to reproduce a previous run.

You need to be logged in to run this model and view results.

Output

{
  "error": "",
  "inferenceTime": 260734,
  "output": "https://media.modelrunner.ai/c1uAbc3JSAuVEqD658NgV.mp4",
  "input": {
    "shift": 5,
    "image_url": "https://media.modelrunner.ai/qf7LUwPh7gD0iVj0lMoCF.jpeg",
    "use_turbo": false,
    "video_url": "https://media.modelrunner.ai/m7GgVU46xQgWaPJGl3kD7.mp4",
    "resolution": "480p",
    "video_quality": "high",
    "guidance_scale": 1,
    "video_write_mode": "balanced",
    "return_frames_zip": false,
    "num_inference_steps": 20
  },
  "logs": "Generated 1 output(s)"
}

Generated in 260.734 seconds

Logs (1 lines)

Examples

Wan 2.2 Animate Move API

Wan 2.2 Animate Move is a video-to-video AI model by wan-video. On ModelRunner it runs through a REST API or via MCP from any AI assistant, at $0.15 per second of video.

POST https://queue.modelrunner.run/wan-video/wan-animate/move

cURL

# Submit a request to the queue. Input fields go at the top level of the
# body. The optional reserved "metadata" object holds your own flat string
# tags — stored on the request, never sent to the model; filter later with
# GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>.
curl -X POST https://queue.modelrunner.run/wan-video/wan-animate/move \
  -H "Authorization: Key $MRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "shift": 5,
    "image_url": "https://media.modelrunner.ai/qf7LUwPh7gD0iVj0lMoCF.jpeg",
    "use_turbo": false,
    "video_url": "https://media.modelrunner.ai/m7GgVU46xQgWaPJGl3kD7.mp4",
    "resolution": "480p",
    "video_quality": "high",
    "guidance_scale": 1,
    "video_write_mode": "balanced",
    "return_frames_zip": false,
    "num_inference_steps": 20,
    "metadata": {
      "project": "my-project"
    }
  }'
# → { "request_id": "...", "status_url": "...", "response_url": "..." }

# Poll status_url until "COMPLETED", then fetch the result
curl "https://queue.modelrunner.run/wan-video/wan-animate/move/requests/$REQUEST_ID/status" \
  -H "Authorization: Key $MRUN_API_KEY"
curl "https://queue.modelrunner.run/wan-video/wan-animate/move/requests/$REQUEST_ID" \
  -H "Authorization: Key $MRUN_API_KEY"

JavaScript

import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("wan-video/wan-animate/move", {
  input: {
    "shift": 5,
    "image_url": "https://media.modelrunner.ai/qf7LUwPh7gD0iVj0lMoCF.jpeg",
    "use_turbo": false,
    "video_url": "https://media.modelrunner.ai/m7GgVU46xQgWaPJGl3kD7.mp4",
    "resolution": "480p",
    "video_quality": "high",
    "guidance_scale": 1,
    "video_write_mode": "balanced",
    "return_frames_zip": false,
    "num_inference_steps": 20
  },
});
console.log(result);

Python

import os
import requests

headers = {"Authorization": f"Key {os.environ['MRUN_API_KEY']}"}

submitted = requests.post(
    "https://queue.modelrunner.run/wan-video/wan-animate/move",
    headers=headers,
    json={
      "shift": 5,
      "image_url": "https://media.modelrunner.ai/qf7LUwPh7gD0iVj0lMoCF.jpeg",
      "use_turbo": false,
      "video_url": "https://media.modelrunner.ai/m7GgVU46xQgWaPJGl3kD7.mp4",
      "resolution": "480p",
      "video_quality": "high",
      "guidance_scale": 1,
      "video_write_mode": "balanced",
      "return_frames_zip": false,
      "num_inference_steps": 20
    },
).json()

# Poll submitted["status_url"] until "COMPLETED", then:
result = requests.get(submitted["response_url"], headers=headers).json()

Input parameters

Name	Type	Required	Description
video_url	string (uri)	yes	URL of the driving video. Its body motion and facial expressions are transferred onto the character image.
image_url	string (uri)	yes	URL of the character image to animate. It is resized and center-cropped to the driving video's aspect ratio.
resolution	enum	no	Output resolution. 480p (default) is the cheapest tier, 580p mid, 720p the highest quality. Default: "480p".
guidance_scale	number	no	Classifier-free guidance strength. Higher values follow the conditioning more strictly; the default of 1 suits most driving videos. Default: 1.
num_inference_steps	integer	no	Number of denoising steps. More steps can add detail at the cost of runtime. Default: 20.
shift	number	no	Flow-matching timestep shift. Leave at the default unless tuning motion fidelity. Default: 5.
video_quality	enum	no	Encoding quality of the returned mp4. Default: "high".
video_write_mode	enum	no	Trade-off used when writing the mp4: 'fast' writes quickest, 'small' produces the smallest file, 'balanced' sits between them. Default: "balanced".
return_frames_zip	boolean	no	Whether to also produce a ZIP archive of the generated frames. Default: false.
use_turbo	boolean	no	Run the accelerated variant: faster generation with a slight reduction in fidelity. Default: false.
seed	integer	no	Random seed. Leave unset for a random result; reuse a value to reproduce a previous run.

Machine-readable: OpenAPI schema · llms.txt

Use Wan 2.2 Animate Move from Claude & Cursor (MCP)

Point Claude Code, Claude Desktop, Cursor, or any MCP client at the ModelRunner MCP server and Wan 2.2 Animate Move becomes a tool your assistant can call directly — it authorizes via OAuth (no API key in config) and runs this model with the run_model tool using the endpoint wan-video/wan-animate/move.

MCP client config (Claude Desktop, Cursor)

{
  "mcpServers": {
    "modelrunner": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.modelrunner.run/mcp"]
    }
  }
}

Claude Code

claude mcp add --transport http modelrunner https://mcp.modelrunner.run/mcp

Then ask your assistant, for example: “Run wan-video/wan-animate/move on ModelRunner to generate video”. MCP setup guide.

Model Details

Wan 2.2 Animate Move turns a still character image into video by transferring a performance from a driving video. Supply one character image and one driving clip: the character reproduces the performer's body motion and facial expression, framed to the driving video's aspect ratio, for as long as that clip runs. There is no text prompt — the whole performance comes from the footage you supply. This is the "Move" mode of Wan Animate: the character is animated in its own scene rather than composited into the driving footage.

## Best for - Animating a character illustration, portrait, or brand mascot from a reference performance - Motion transfer — reusing one filmed performance across several different characters - Expression and lip-movement transfer for talking-head clips, driven by video rather than a script - Previsualizing choreography, gestures, or presenter delivery with a stand-in character - Bulk character animation where a written prompt would be too vague to direct the motion

## Choose another model when - You have no driving video and want motion invented from a description — use a text-to-video or image-to-video model - You want to restyle a clip or change its background from a written instruction — use a prompt-driven video editing model - You need mouth movement driven by an audio track instead of by video — use a lip-sync model - You need a still image rather than motion — use an image-editing or text-to-image model

## Tips - Use a clear, unobstructed character image; it is center-cropped, so leave margin around the subject - Keep the driving clip short — cost scales with output duration - `resolution` sets both quality and price: `480p` (the default) is cheapest, then `580p`, then `720p` - `use_turbo` runs an accelerated variant — faster, with a small fidelity trade-off; leave it off for best quality - Framing mismatch between the image and the driving performer is the main cause of weak results

## Advanced Configuration - `video_quality` (`low`, `medium`, `high`, `maximum`; default `high`) and `video_write_mode` (`fast`, `balanced`, `small`; default `balanced`) set how the mp4 is encoded — use `small` for a lighter file and `maximum` for downstream re-encoding.

## Limitations - Large differences in body proportion, camera distance, or framing between the character image and the driving performer can cause warping - Fast motion, heavy occlusion, and hands are the weakest cases - Duration and frame rate follow the driving clip and cannot be set directly

To run via the ModelRunner JavaScript client: ```js import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("wan-video/wan-animate/move", { input: { image_url: "https://media.modelrunner.ai/qf7LUwPh7gD0iVj0lMoCF.jpeg", video_url: "https://media.modelrunner.ai/m7GgVU46xQgWaPJGl3kD7.mp4", resolution: "480p", }, }); ```

Wan 2.2 Animate Move API

Model Input

Input

Additional Settings

Model Output

Output

Model Example Requests

Examples

Wan 2.2 Animate Move API

cURL

JavaScript

Python

Input parameters

Use Wan 2.2 Animate Move from Claude & Cursor (MCP)

MCP client config (Claude Desktop, Cursor)

Claude Code

Model Details

Model Details