Name: Qwen Image Edit Plus Fusion Product Background API
Brand: magicapi
SKU: api-magicapi-image-background-fusion-qwen-edit-apps-qwen-image-edit-plus-lora-fusion
Availability: InStock
Author: API Market

Shashank Agarwal

# 🧩 Product Background Fusion API - Perspective and Light Matching (Qwen Image Edit Plus LoRA)

## Overview
Fusion is an image editing model built on Qwen-Image-Edit-Plus with a Fusion LoRA adapter. Its developer describes it as product and object blending that corrects perspective and lighting so the subject fits naturally into a new background.

Send a photo of a product or object as `image`. The `prompt` should start with a Chinese trigger phrase (shown below); after it, describe in English the scene and lighting you want around the subject. The default prompt already contains the trigger phrase, so `image` is the only required field. Each request returns a list with the URL of one image.

## Features
- **Background fusion**: places the subject of your image into the scene you describe, with perspective and lighting adjusted to match.
- **Trigger phrase**: start the prompt with `溶图,纠正产品透视角度和光影并使产品融入背景` (meaning "blend the product into the background with corrected perspective and lighting"), then continue in English.
- **Aspect ratio**: `match_input_image` (default) keeps the shape of your image; you can also choose `1:1`, `16:9`, `9:16`, `4:3` or `3:4`.
- **Speed or detail**: `go_fast` set to `true` runs an 8-step preset, `false` runs a 40-step preset. Set `num_inference_steps` (1 to 40) to choose the step count yourself.
- **LoRA controls**: `lora_scale` (0 to 4) sets the strength of the Fusion adapter; `lora_weights` swaps in different LoRA weights, such as your own LoRA file.
- **Guidance**: `true_guidance_scale` (0 to 10, default 1).
- **Output format**: `webp` (default), `jpg` or `png`, with `output_quality` from 0 to 100 for lossy formats.
- **Seed**: set `seed` to reproduce a result.

## Use Cases
1. **Product listings**: show a product photo in a styled room, on a desk or outdoors without a new photo shoot.
2. **Lifestyle marketing images**: place one product into several different scenes for ads and social posts.
3. **Concept mockups**: preview how an object looks in a planned setting before staging it for real.
4. **Catalog variety**: create several backgrounds for the same item from one product photo.

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

![Example input: an astronaut figure](https://blog.api.market/wp-content/uploads/2026/09/kling-v2-6-motion-control-input-character-api-market.webp)

Prompt: "溶图,纠正产品透视角度和光影并使产品融入背景. Place the astronaut figure on a walnut desk beside a sunny window in a cozy home studio, with soft morning light and a natural shadow under its boots."

All other fields were left at their defaults (`match_input_image`, `webp`).

### Example Output
![Fusion output: the astronaut figure from the input placed in the desk scene described in the prompt](https://blog.api.market/wp-content/uploads/2026/10/qwen-image-edit-plus-lora-fusion-output-api-market.webp)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `image` | string (URL) | Yes |  |  | Image URL (jpeg, png, gif or webp). |
| `prompt` | string | No | `溶图,纠正产品透视角度和光影并使产品融入背景, 描述新的环境与灯光让主体自然融入. Blend the product into a modern showroom with warm diffused lighting.` |  | Prompt hint: paste the Chinese trigger `溶图,纠正产品透视角度和光影并使产品融入背景` (it means "blend the product into the background with corrected perspective and lighting"), then describe in English the scene and lighting you want. |
| `aspect_ratio` | string | No | `match_input_image` | `match_input_image`, `1:1`, `16:9`, `9:16`, `4:3`, `3:4` | Aspect ratio for the generated image |
| `go_fast` | boolean | No | none |  | If num_inference_steps is omitted, true runs the 8-step Lightning preset and false runs the 40-step detailed preset. |
| `num_inference_steps` | integer | No | none | 1 to 40 | Explicit denoising step count (1-40). Leave blank to use the go_fast presets (fast = 8 steps, slow = 40 steps). |
| `lora_weights` | string | No | `""` |  | LoRA weights to apply. Leave blank to use the default Fusion adapter, or point it at your own LoRA file (for example a public huggingface.co URL). |
| `lora_scale` | number | No | none | 0 to 4 | Strength applied to the selected LoRA. Leave blank for the default strength. |
| `true_guidance_scale` | number | No | `1` | 0 to 10 | True classifier-free guidance scale passed to the pipeline. |
| `seed` | integer | No | none |  | Random seed. Set for reproducible generation. |
| `output_format` | string | No | `webp` | `webp`, `jpg`, `png` | Format of the output images. |
| `output_quality` | integer | No | `95` | 0 to 100 | Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs. |
| `disable_safety_checker` | boolean | No | `false` |  | Disable safety checker for generated images. |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.038 per API unit.
- **1 API unit = 1 output image.** Each request returns one image (a one-item list), so each successful request costs $0.038.
- Every input setting (aspect ratio, step count, output format) is billed the same. The example request above cost $0.038.
- **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/image-background-fusion-qwen-edit-apps-qwen-image-edit-plus-lora-fusion`

### 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 1 second 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/image-background-fusion-qwen-edit-apps-qwen-image-edit-plus-lora-fusion/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "0fdb72e0360d0b91ac1e8d4b5ea6968042c316a4e9ae3e4931a6fa56bec36891",
    "input": {
      "image": "https://blog.api.market/wp-content/uploads/2026/09/kling-v2-6-motion-control-input-character-api-market.webp",
      "prompt": "溶图,纠正产品透视角度和光影并使产品融入背景. Place the astronaut figure on a walnut desk beside a sunny window in a cozy home studio, with soft morning light and a natural shadow under its boots."
    }
  }'
```

Response (201 Created):
```json
{
  "id": "f537x8r12xrmw0d17g4b65vh1g",
  "version": "0fdb72e0360d0b91ac1e8d4b5ea6968042c316a4e9ae3e4931a6fa56bec36891",
  "status": "starting",
  "created_at": "2026-10-14T20:08:40.066Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/image-background-fusion-qwen-edit-apps-qwen-image-edit-plus-lora-fusion/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "f537x8r12xrmw0d17g4b65vh1g",
  "status": "succeeded",
  "output": [
    "https://blog.api.market/wp-content/uploads/2026/10/qwen-image-edit-plus-lora-fusion-output-api-market.webp"
  ],
  "error": null,
  "created_at": "2026-10-14T20:08:40.066Z",
  "started_at": "2026-10-14T20:08:41.066Z",
  "completed_at": "2026-10-14T20:08:55.066Z"
}
```

`output` is a list of URL strings. It holds one generated image, in the format set by `output_format`.

### 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 `image` or a `lora_scale` outside 0 to 4.
- **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/image-background-fusion-qwen-edit-apps-qwen-image-edit-plus-lora-fusion";

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: "0fdb72e0360d0b91ac1e8d4b5ea6968042c316a4e9ae3e4931a6fa56bec36891", 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 = 1000;
  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({
  "image": "https://blog.api.market/wp-content/uploads/2026/09/kling-v2-6-motion-control-input-character-api-market.webp",
  "prompt": "溶图,纠正产品透视角度和光影并使产品融入背景. Place the astronaut figure on a walnut desk beside a sunny window in a cozy home studio, with soft morning light and a natural shadow under its boots."
}).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/image-background-fusion-qwen-edit-apps-qwen-image-edit-plus-lora-fusion"

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": "0fdb72e0360d0b91ac1e8d4b5ea6968042c316a4e9ae3e4931a6fa56bec36891", "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 = 1
    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({
  "image": "https://blog.api.market/wp-content/uploads/2026/09/kling-v2-6-motion-control-input-character-api-market.webp",
  "prompt": "溶图,纠正产品透视角度和光影并使产品融入背景. Place the astronaut figure on a walnut desk beside a sunny window in a cozy home studio, with soft morning light and a natural shadow under its boots."
}))
```

## FAQ
- **Why does the prompt start with Chinese text?**
  - The Fusion adapter is triggered by the phrase `溶图,纠正产品透视角度和光影并使产品融入背景` ("blend the product into the background with corrected perspective and lighting"). Paste it at the start of your prompt and describe the new scene in English. If you leave `prompt` out, the default prompt already includes it and asks for a modern showroom with warm diffused lighting.
- **Does it only work with products?**
  - The developer describes it for product and object blending. The example above uses a figurine rather than a product photo.
- **How do I trade speed for detail?**
  - Leave `num_inference_steps` empty and set `go_fast` to `true` for the 8-step preset or `false` for the 40-step preset. Or set `num_inference_steps` (1 to 40) yourself.
- **Will the output keep the shape of my image?**
  - Yes, by default. `aspect_ratio` is `match_input_image` unless you choose `1:1`, `16:9`, `9:16`, `4:3` or `3:4`.
- **How many images does one request return, and what does it cost?**
  - One image, returned as a one-item list in `output`. It costs $0.038.
- **Is the 201 response from POST an error?**
  - No. 201 Created means the prediction was accepted. Use the returned `id` to poll GET `/predictions/{id}` until `status` is `succeeded`, `failed` or `canceled`.

## Contact Us

If you have any questions or need assistance, you can reach out to us through the embedded chat on the bottom right corner of the page or send an email to hello@magicapi.com.


Blend product and object photos into new backgrounds with corrected perspective and lighting using a Fusion LoRA on Qwen-Image-Edit-Plus. $0.038 per image.

Qwen Image Edit Plus Fusion Product Background API