# Wan 2.2 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.

## Overview

- **Endpoint**: `https://queue.modelrunner.run/wan-video/wan-animate/move`
- **Model ID**: `wan-video/wan-animate/move`
- **Category**: video-to-video
- **Kind**: inference
- **Tags**: wan, wan-animate, alibaba, video-to-video, motion-transfer, character-animation

## Pricing

- **Price**: $0.15 per output second

## Request Lifecycle

This model runs on the ModelRunner **asynchronous queue API** — a single POST does not return the output. Every call requires an `Authorization: Key $MODEL_RUNNER_KEY` header. Run three steps:

1. **Submit** — `POST https://queue.modelrunner.run/wan-video/wan-animate/move` with a JSON body holding the input fields at the top level. The body may also include a reserved top-level `metadata` object — a flat string map (max 16 keys, key ≤64 / value ≤512 chars) stored on the request for your own tagging. It is never sent to the model; filter your request history with `GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>` (exact key=value matches, AND-ed). The response carries request handles only (no output yet):

```json
{
  "status": "IN_QUEUE",
  "request_id": "<21-char id>",
  "status_url": "https://queue.modelrunner.run/wan-video/wan-animate/move/requests/<request_id>/status",
  "response_url": "https://queue.modelrunner.run/wan-video/wan-animate/move/requests/<request_id>",
  "cancel_url": "https://queue.modelrunner.run/wan-video/wan-animate/move/requests/<request_id>/cancel"
}
```

2. **Poll status** — `GET <status_url>` until `status` is `COMPLETED`. Possible values are `IN_QUEUE`, `IN_PROGRESS`, `COMPLETED`, `FAILED`, `CANCELLED`. A `FAILED` request responds with HTTP 400 and an `error` field.

3. **Read result** — `GET <response_url>`. Returns the finished request, including the generated `output`:

```json
{ "id": "<request_id>", "status": "COMPLETED", "output": ..., "input": ... }
```

The JavaScript and Python SDKs below perform steps 2–3 for you. In any language without an SDK (Swift, Go, Kotlin, etc.) you must implement the polling loop and the final result fetch yourself — see the cURL example for the full flow.

### Input Schema

- **`seed`** (`integer | null`, _optional_):
  Random seed. Leave unset for a random result; reuse a value to reproduce a previous run.

- **`shift`** (`number`, _optional_):
  Flow-matching timestep shift. Leave at the default unless tuning motion fidelity.
  - Default: `5`
  - Range: `1` to `10`

- **`image_url`** (`string`, _required_):
  URL of the character image to animate. It is resized and center-cropped to the driving video's aspect ratio.

- **`use_turbo`** (`boolean`, _optional_):
  Run the accelerated variant: faster generation with a slight reduction in fidelity.
  - Default: `false`

- **`video_url`** (`string`, _required_):
  URL of the driving video. Its body motion and facial expressions are transferred onto the character image.

- **`resolution`** (`ResolutionEnum`, _optional_):
  Output resolution. 480p (default) is the cheapest tier, 580p mid, 720p the highest quality.
  - Default: `"480p"`
  - Options: `"480p"`, `"580p"`, `"720p"`

- **`video_quality`** (`VideoQualityEnum`, _optional_):
  Encoding quality of the returned mp4.
  - Default: `"high"`
  - Options: `"low"`, `"medium"`, `"high"`, `"maximum"`

- **`guidance_scale`** (`number`, _optional_):
  Classifier-free guidance strength. Higher values follow the conditioning more strictly; the default of 1 suits most driving videos.
  - Default: `1`
  - Range: `1` to `10`

- **`video_write_mode`** (`VideoWriteModeEnum`, _optional_):
  Trade-off used when writing the mp4: 'fast' writes quickest, 'small' produces the smallest file, 'balanced' sits between them.
  - Default: `"balanced"`
  - Options: `"fast"`, `"balanced"`, `"small"`

- **`return_frames_zip`** (`boolean`, _optional_):
  Whether to also produce a ZIP archive of the generated frames.
  - Default: `false`

- **`num_inference_steps`** (`integer`, _optional_):
  Number of denoising steps. More steps can add detail at the cost of runtime.
  - Default: `20`
  - Range: `2` to `40`

### Output Schema

_No `Output` schema properties are available._

## Default Example

**Input**

```json
{
  "shift": 5,
  "image_url": "https://media.modelrunner.ai/rI1wboXaj1eKII7S1ld5k.jpeg",
  "use_turbo": false,
  "video_url": "https://media.modelrunner.ai/xCu3Uic8zYdc9dHfxmTCU.mp4",
  "resolution": "480p",
  "video_quality": "high",
  "guidance_scale": 1,
  "video_write_mode": "balanced",
  "return_frames_zip": false,
  "num_inference_steps": 20
}
```

**Output**

```json
"https://media.modelrunner.ai/Ov6fvgvvzhJyD3G6JVjmv.mp4"
```
## Usage Examples

### cURL

The queue API is asynchronous: submit the request, poll `status_url` until it is `COMPLETED`, then read the result from `response_url`. Requires `jq`.

```bash
# 1. Submit the request (returns request handles, not the output)
SUBMIT=$(curl --silent --request POST \
  --url https://queue.modelrunner.run/wan-video/wan-animate/move \
  --header "Authorization: Key $MODEL_RUNNER_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "shift": 5,
  "image_url": "https://media.modelrunner.ai/rI1wboXaj1eKII7S1ld5k.jpeg",
  "use_turbo": false,
  "video_url": "https://media.modelrunner.ai/xCu3Uic8zYdc9dHfxmTCU.mp4",
  "resolution": "480p",
  "video_quality": "high",
  "guidance_scale": 1,
  "video_write_mode": "balanced",
  "return_frames_zip": false,
  "num_inference_steps": 20
}')

STATUS_URL=$(echo "$SUBMIT" | jq -r '.status_url')
RESPONSE_URL=$(echo "$SUBMIT" | jq -r '.response_url')

# 2. Poll until the request leaves the queue / in-progress state
while true; do
  STATUS=$(curl --silent --url "$STATUS_URL" \
    --header "Authorization: Key $MODEL_RUNNER_KEY" | jq -r '.status')
  echo "Status: $STATUS"
  case "$STATUS" in
    COMPLETED) break ;;
    FAILED|CANCELLED) echo "Request $STATUS"; exit 1 ;;
  esac
  sleep 1
done

# 3. Read the finished request, including the generated output
curl --silent --url "$RESPONSE_URL" \
  --header "Authorization: Key $MODEL_RUNNER_KEY"
```

### JavaScript

```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/rI1wboXaj1eKII7S1ld5k.jpeg",
  "use_turbo": false,
  "video_url": "https://media.modelrunner.ai/xCu3Uic8zYdc9dHfxmTCU.mp4",
  "resolution": "480p",
  "video_quality": "high",
  "guidance_scale": 1,
  "video_write_mode": "balanced",
  "return_frames_zip": false,
  "num_inference_steps": 20
}
});

console.log(result.data);
```

### Python

```python
import asyncio
import modelrunner_ai

async def main():
    response = await modelrunner_ai.submit_async(
        "wan-video/wan-animate/move",
        arguments={
  "shift": 5,
  "image_url": "https://media.modelrunner.ai/rI1wboXaj1eKII7S1ld5k.jpeg",
  "use_turbo": false,
  "video_url": "https://media.modelrunner.ai/xCu3Uic8zYdc9dHfxmTCU.mp4",
  "resolution": "480p",
  "video_quality": "high",
  "guidance_scale": 1,
  "video_write_mode": "balanced",
  "return_frames_zip": false,
  "num_inference_steps": 20
}
    )
    result = await response.get()
    print(result["output"])

asyncio.run(main())
```

## Additional Resources

- [Playground](https://modelrunner.ai/models/wan-video/wan-animate/move)
- [OpenAPI Schema](https://modelrunner.ai/models/wan-video/wan-animate/move/openapi.json)
- [LLM Instructions](https://modelrunner.ai/models/wan-video/wan-animate/move/llms.txt)