Name: Wan 2.2 First and Last Frame API: Keyframe to Video
Brand: magicapi
SKU: api-magicapi-first-and-last-frame-video-ai-lucataco-wan-2-2-first-last-frame
Availability: InStock
Author: API Market

Shashank Agarwal

# 🎬 First and Last Frame to Video API (Alibaba Wan 2.2)

## Overview
Wan 2.2 First and Last Frame makes a video that starts on one image and ends on another. Send a `start_image`, an `end_image` and a `prompt` describing the transition, and get back the URL of an MP4 video.

This version runs Wan 2.2 with a Lightning LoRA for 8-step inference. The video length is set with `duration_seconds` (0.5 to 10 seconds, default 5).

## Features
- **Start and end frames**: `start_image` and `end_image` are both required.
- **Transition prompt**: `prompt` describes what happens between the frames (default `animate`).
- **Duration**: `duration_seconds` from 0.5 to 10 (default 5).
- **Fast inference**: 8 steps by default (`num_inference_steps` 1 to 30).
- **Two guidance scales**: `guidance_scale` for high noise and `guidance_scale_2` for low noise (0 to 10, default 1), plus scheduler `shift` (1 to 10, default 8).
- **Negative prompt and seed** for control and reproducibility.

## Use Cases
1. **Transformations**: animate a change from one state of a subject to another.
2. **Product reveals**: move from a packaging shot to the product in use.
3. **Storyboards**: turn two key frames into a motion test.
4. **Loops**: use the same image as start and end for a looping clip.

## Visual Examples
### Example Input
Start image:

![Wan 2.2 First and Last Frame example start image](https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-2-api-market.jpeg)

End image:

![Wan 2.2 First and Last Frame example end image](https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-api-market.webp)

Prompt: "the action figure turns into a caped superhero"

`duration_seconds`: `5`, `num_inference_steps`: `8`, `guidance_scale` and `guidance_scale_2`: `1`, `shift`: `8`, `seed`: `0`, default negative prompt.

### Example Output
[Watch the example output](https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-output-0dgms5zf-api-market.mp4)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `start_image` | string (URL) | Yes | none |  | Start frame image (URL) |
| `end_image` | string (URL) | Yes | none |  | End frame image (URL) |
| `prompt` | string | No | `animate` |  | Prompt describing the transition between images |
| `negative_prompt` | string | No | `色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走,过曝，` |  | Negative prompt |
| `duration_seconds` | number | No | `5` | `0.5` to `10` | Video duration in seconds |
| `num_inference_steps` | integer | No | `8` | `1` to `30` | Number of inference steps |
| `guidance_scale` | number | No | `1` | `0` to `10` | Guidance scale for high noise |
| `guidance_scale_2` | number | No | `1` | `0` to `10` | Guidance scale for low noise |
| `shift` | number | No | `8` | `1` to `10` | Scheduler shift parameter |
| `seed` | integer | No | none |  | Random seed for reproducibility. Leave empty for a random seed |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.098 per API unit.
- **1 API unit = 1 request with up to 57 seconds of processing.** A request that takes longer uses one unit per started block of 57 seconds (for example, 120 seconds of processing uses 3 units, $0.294).
- Longer videos and more inference steps take longer and can use more units.
- **Example cost**: the example request took 21.9 seconds of processing, which is 1 unit, $0.098.
- **Failed or cancelled predictions are not billed.**
- **GET requests to check prediction status** are free and don't count toward API unit usage.
- **No monthly subscription fee.**

## API Documentation

### Base URL
All API requests go to: `https://prod.api.market/api/v1/magicapi/first-and-last-frame-video-ai-lucataco-wan-2-2-first-last-frame`

### Authentication
Send your api.market API key in the `x-api-market-key` header:
```
x-api-market-key: YOUR_API_MARKET_KEY
```

### Asynchronous Processing
**IMPORTANT**: This API is asynchronous. The flow is:
1. Send a POST request to `/predictions`. A successful request returns **201 Created** with a prediction `id` and a `status` of `starting` or `processing`.
2. Poll GET `/predictions/{id}` with that ID.
3. Keep polling while `status` is `starting` or `processing`.
4. Stop when `status` is `succeeded` (read `output`), `failed` or `canceled` (read `error`).

A polling interval of 2 seconds and an overall timeout of 600 seconds (10 minutes) work well.

### Endpoints

#### Create a Prediction
```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/first-and-last-frame-video-ai-lucataco-wan-2-2-first-last-frame/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
      "version": "003fd8a38ff17cb6022c3117bb90f7403cb632062ba2b098710738d116847d57",
      "input": {
        "seed": 0,
        "shift": 8,
        "prompt": "the action figure turns into a caped superhero",
        "end_image": "https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-api-market.webp",
        "start_image": "https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-2-api-market.jpeg",
        "guidance_scale": 1,
        "negative_prompt": "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走,过曝，",
        "duration_seconds": 5,
        "guidance_scale_2": 1,
        "num_inference_steps": 8
      }
    }'
```

Response (201 Created):
```json
{
  "id": "0dgms5zfjnrp40d1d268x5gc1m",
  "version": "003fd8a38ff17cb6022c3117bb90f7403cb632062ba2b098710738d116847d57",
  "status": "starting",
  "created_at": "2026-10-23T11:37:09.529Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/first-and-last-frame-video-ai-lucataco-wan-2-2-first-last-frame/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "0dgms5zfjnrp40d1d268x5gc1m",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-output-0dgms5zf-api-market.mp4",
  "error": null,
  "created_at": "2026-10-23T11:37:09.529Z",
  "started_at": "2026-10-23T11:39:26.020Z",
  "completed_at": "2026-10-23T11:39:47.945Z"
}
```

`output` is a single URL string pointing to the MP4 video.

### Status Codes
- **201 Created**: prediction created (POST `/predictions`). This is the normal success response.
- **200 OK**: prediction details returned (GET `/predictions/{id}`).
- **400 Bad Request**: invalid input, for example a missing `start_image` or `end_image`, or a `duration_seconds` above 10.
- **401 Unauthorized**: missing or invalid API key.
- **404 Not Found**: no prediction with that ID.
- **429 Too Many Requests**: rate limit exceeded.
- **500 Internal Server Error** or **503 Service Unavailable**: temporary server problem; retry later.

## Rate Limits
All plans allow 10 requests per second. Requests above that return `429 Too Many Requests`; back off before retrying and honour any `Retry-After` header. Polling a prediction with GET is free but still counts toward the per-second limit, so poll every 1 to 5 seconds rather than in a tight loop.

## Error Handling
- Check the HTTP status code of every request and read the response body for the error message.
- A prediction can also end with `status` set to `failed`; the `error` field then explains why. Failed predictions are not billed.
- Retry 429, 500 and 503 responses with exponential backoff. Fix the input before retrying a 400.

## Implementation Guide
### Example Code

#### JavaScript
```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/first-and-last-frame-video-ai-lucataco-wan-2-2-first-last-frame";

async function createAndWaitForPrediction(input) {
  const response = await fetch(`${baseUrl}/predictions`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "x-api-market-key": apiKey
    },
    body: JSON.stringify({ version: "003fd8a38ff17cb6022c3117bb90f7403cb632062ba2b098710738d116847d57", input })
  });

  if (response.status !== 201) {
    throw new Error(`API error: ${response.status} ${await response.text()}`);
  }

  const prediction = await response.json();
  return pollForResults(prediction.id);
}

const delay = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function pollForResults(predictionId) {
  const maxTimeout = 600000;
  const pollInterval = 2000;
  const startTime = Date.now();

  while (Date.now() - startTime < maxTimeout) {
    const response = await fetch(`${baseUrl}/predictions/${predictionId}`, {
      headers: { "x-api-market-key": apiKey }
    });

    if (!response.ok) {
      throw new Error(`API error: ${response.status} ${await response.text()}`);
    }

    const prediction = await response.json();

    if (prediction.status === "succeeded") {
      return prediction.output;
    }

    if (prediction.status === "failed" || prediction.status === "canceled") {
      throw new Error(`Prediction ${prediction.status}: ${prediction.error || "Unknown error"}`);
    }

    await delay(pollInterval);
  }

  throw new Error("Prediction timed out after 10 minutes");
}

createAndWaitForPrediction({
    "seed": 0,
    "shift": 8,
    "prompt": "the action figure turns into a caped superhero",
    "end_image": "https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-api-market.webp",
    "start_image": "https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-2-api-market.jpeg",
    "guidance_scale": 1,
    "negative_prompt": "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走,过曝，",
    "duration_seconds": 5,
    "guidance_scale_2": 1,
    "num_inference_steps": 8
  }).then((output) => console.log(output));
```

#### Python
```python
import time
import requests

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/first-and-last-frame-video-ai-lucataco-wan-2-2-first-last-frame"


def create_and_wait_for_prediction(input_data):
    headers = {"Content-Type": "application/json", "x-api-market-key": API_KEY}
    response = requests.post(
        f"{BASE_URL}/predictions",
        headers=headers,
        json={"version": "003fd8a38ff17cb6022c3117bb90f7403cb632062ba2b098710738d116847d57", "input": input_data},
    )

    if response.status_code != 201:
        raise Exception(f"API error: {response.status_code} {response.text}")

    return poll_for_results(response.json()["id"])


def poll_for_results(prediction_id):
    max_timeout = 600
    poll_interval = 2
    headers = {"x-api-market-key": API_KEY}
    start_time = time.time()

    while time.time() - start_time < max_timeout:
        response = requests.get(f"{BASE_URL}/predictions/{prediction_id}", headers=headers)

        if response.status_code != 200:
            raise Exception(f"API error: {response.status_code} {response.text}")

        prediction = response.json()

        if prediction["status"] == "succeeded":
            return prediction["output"]

        if prediction["status"] in ("failed", "canceled"):
            raise Exception(f"Prediction {prediction['status']}: {prediction.get('error') or 'Unknown error'}")

        time.sleep(poll_interval)

    raise Exception("Prediction timed out after 10 minutes")


print(create_and_wait_for_prediction({
    "seed": 0,
    "shift": 8,
    "prompt": "the action figure turns into a caped superhero",
    "end_image": "https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-api-market.webp",
    "start_image": "https://blog.api.market/wp-content/uploads/2026/10/wan-2-2-first-last-frame-input-2-api-market.jpeg",
    "guidance_scale": 1,
    "negative_prompt": "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走,过曝，",
    "duration_seconds": 5,
    "guidance_scale_2": 1,
    "num_inference_steps": 8
}))
```

## FAQ
- **Do I need both images?**
  - Yes. `start_image` and `end_image` are required. The video starts on the first and ends on the second.
- **How long can the video be?**
  - From 0.5 to 10 seconds, set with `duration_seconds` (default 5).
- **What should the prompt say?**
  - Describe the motion or change between the two frames. The default prompt is `animate`.
- **Can I make a loop?**
  - Send the same image as `start_image` and `end_image` and describe the motion in between.
- **What does one request cost?**
  - One unit ($0.098) covers up to 57 seconds of processing. The example (5 seconds, 8 steps) took 21.9 seconds, so it cost $0.098.

## Contact Us
If you have any questions, reach out to us through the chat on the bottom right of the page. You can also email our support team at hello@magicapi.com.


Turn a start frame and an end frame into a video of up to 10 seconds with Wan 2.2 and 8-step inference. Pay as you go, no monthly fee.