Name: SUPIR-v0Q API: AI Photo Restoration and Upscaling
Brand: magicapi
SKU: api-magicapi-photo-realistic-image-restoration-api-cjwbw-supir-v0q
Availability: InStock
Author: API Market

Shashank Agarwal

# 🖼️ Photo Restoration API (cjwbw SUPIR-v0Q)

## Overview
SUPIR is a photo restoration model. Its developers describe it as practicing model scaling for photo-realistic image restoration in the wild. This listing runs the SUPIR-v0Q checkpoint, published by cjwbw. It does not caption the image with a separate vision-language model; the restoration is guided by the `a_prompt` and `n_prompt` text (or their defaults).

Send a low-quality `image`. Set `upscale` to enlarge it and `min_size` for the minimum output resolution. Each request returns the URL of one PNG image.

## Features
- **Photo restoration**: recover detail in blurry, noisy or compressed photos.
- **Upscaling**: `upscale` sets the upsampling ratio (default 1, no enlargement) and `min_size` the minimum output resolution (default 1024).
- **Prompt guidance**: `a_prompt` describes the look you want and `n_prompt` what to avoid; both have defaults.
- **Color fixing**: `color_fix_type` `Wavelet` (default), `AdaIn` or `None` keeps colors close to the input.
- **Sampler controls**: `edm_steps` (1 to 500), `s_cfg` (1 to 20), stage strengths `s_stage1` and `s_stage2`, and EDM `s_churn` and `s_noise`.
- **Reproducible output**: set `seed` to repeat a result.

## Use Cases
1. **Old and damaged photos**: bring back detail in scanned prints and family photos.
2. **Low-resolution images**: enlarge small product or profile photos with `upscale`.
3. **Compressed images**: clean up JPEG artifacts and noise in images saved at low quality.
4. **Archive digitization**: improve batches of scanned images before publishing them.

## Visual Examples
### Example Input
A low-quality photo with `upscale` 1, `min_size` 1024, 50 EDM steps, `color_fix_type` `Wavelet` and the default prompts:

![Input image: the low-quality photo sent for restoration](https://blog.api.market/wp-content/uploads/2026/10/supir-v0q-input-api-market.png)

```json
{
  "image": "https://blog.api.market/wp-content/uploads/2026/10/supir-v0q-input-api-market.png",
  "s_cfg": 7.5,
  "s_churn": 5,
  "s_noise": 1.003,
  "upscale": 1,
  "a_prompt": "Cinematic, High Contrast, highly detailed, taken using a Canon EOS R camera, hyper detailed photo - realistic maximum detail, 32k, Color Grading, ultra HD, extreme meticulous detailing, skin pore detailing, hyper sharpness, perfect without deformations.",
  "min_size": 1024,
  "n_prompt": "painting, oil painting, illustration, drawing, art, sketch, oil painting, cartoon, CG Style, 3D render, unreal engine, blurring, dirty, messy, worst quality, low quality, frames, watermark, signature, jpeg artifacts, deformed, lowres, over-smooth",
  "s_stage1": -1,
  "s_stage2": 1,
  "edm_steps": 50,
  "linear_CFG": false,
  "color_fix_type": "Wavelet",
  "spt_linear_CFG": 1,
  "linear_s_stage2": false,
  "spt_linear_s_stage2": 0
}
```

### Example Output
![SUPIR-v0Q output: the restored, sharper version of the input photo](https://blog.api.market/wp-content/uploads/2026/10/supir-v0q-output-ttw01wxa-api-market.png)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `image` | string (URL) | Yes |  |  | Low quality input image URL |
| `upscale` | integer | No | `1` |  | Upsampling ratio of the input image |
| `min_size` | number | No | `1024` |  | Minimum resolution of the output image |
| `edm_steps` | integer | No | `50` | 1 to 500 | Number of steps for the EDM sampling schedule |
| `a_prompt` | string | No | see below the table |  | Positive prompt added to the restoration. The default describes a sharp, highly detailed photo. |
| `n_prompt` | string | No | see below the table |  | Negative prompt for the restoration. The default lists painting, cartoon, blur and low-quality artifacts to avoid. |
| `color_fix_type` | string | No | `Wavelet` | `None`, `AdaIn`, `Wavelet` | Color fixing method |
| `s_stage1` | integer | No | `-1` |  | Control strength of stage 1 (a negative value turns it off) |
| `s_churn` | number | No | `5` |  | Churn hyperparameter of the EDM sampler |
| `s_noise` | number | No | `1.003` |  | Noise hyperparameter of the EDM sampler |
| `s_cfg` | number | No | `7.5` | 1 to 20 | Classifier-free guidance scale for the prompts |
| `s_stage2` | number | No | `1` |  | Control strength of stage 2 |
| `linear_CFG` | boolean | No | `false` |  | Linearly (with sigma) increase CFG from spt_linear_CFG to s_cfg |
| `linear_s_stage2` | boolean | No | `false` |  | Linearly (with sigma) increase s_stage2 from spt_linear_s_stage2 to s_stage2 |
| `spt_linear_CFG` | number | No | `1` |  | Start point of the linear CFG increase |
| `spt_linear_s_stage2` | number | No | `0` |  | Start point of the linear s_stage2 increase |
| `seed` | integer | No | none |  | Random seed. Leave blank to randomize the seed |

Default `a_prompt`: "Cinematic, High Contrast, highly detailed, taken using a Canon EOS R camera, hyper detailed photo - realistic maximum detail, 32k, Color Grading, ultra HD, extreme meticulous detailing, skin pore detailing, hyper sharpness, perfect without deformations."

Default `n_prompt`: "painting, oil painting, illustration, drawing, art, sketch, oil painting, cartoon, CG Style, 3D render, unreal engine, blurring, dirty, messy, worst quality, low quality, frames, watermark, signature, jpeg artifacts, deformed, lowres, over-smooth"

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.048 per API unit.
- **1 API unit = 1 request with up to 44 seconds of processing time.** A request that finishes within 44 seconds costs one unit ($0.048).
- A request that runs longer bills one more unit for each started block of 44 seconds: more than 44 and up to 88 seconds is 2 units ($0.096), up to 132 seconds is 3 units ($0.144), and so on.
- **Example cost**: the example request took 13.2 seconds of processing, which is 1 unit = **$0.048**.
- Larger `upscale` and `min_size` values and more `edm_steps` take longer and can use more units.
- **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/photo-realistic-image-restoration-api-cjwbw-supir-v0q`

### 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/photo-realistic-image-restoration-api-cjwbw-supir-v0q/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "ede69f6a5ae7d09f769d683347325b08d2f83a93d136ed89747941205e0a71da",
    "input": {
      "image": "https://blog.api.market/wp-content/uploads/2026/10/supir-v0q-input-api-market.png",
      "s_cfg": 7.5,
      "s_churn": 5,
      "s_noise": 1.003,
      "upscale": 1,
      "a_prompt": "Cinematic, High Contrast, highly detailed, taken using a Canon EOS R camera, hyper detailed photo - realistic maximum detail, 32k, Color Grading, ultra HD, extreme meticulous detailing, skin pore detailing, hyper sharpness, perfect without deformations.",
      "min_size": 1024,
      "n_prompt": "painting, oil painting, illustration, drawing, art, sketch, oil painting, cartoon, CG Style, 3D render, unreal engine, blurring, dirty, messy, worst quality, low quality, frames, watermark, signature, jpeg artifacts, deformed, lowres, over-smooth",
      "s_stage1": -1,
      "s_stage2": 1,
      "edm_steps": 50,
      "linear_CFG": false,
      "color_fix_type": "Wavelet",
      "spt_linear_CFG": 1,
      "linear_s_stage2": false,
      "spt_linear_s_stage2": 0
    }
  }'
```

Response (201 Created):
```json
{
  "id": "ttw01wxapsrmw0d1d0ttqt2j6c",
  "version": "ede69f6a5ae7d09f769d683347325b08d2f83a93d136ed89747941205e0a71da",
  "status": "starting",
  "created_at": "2026-10-23T10:01:50.166Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/photo-realistic-image-restoration-api-cjwbw-supir-v0q/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "ttw01wxapsrmw0d1d0ttqt2j6c",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/10/supir-v0q-output-ttw01wxa-api-market.png",
  "error": null,
  "created_at": "2026-10-23T10:01:50.166Z",
  "started_at": "2026-10-23T10:04:10.004Z",
  "completed_at": "2026-10-23T10:04:23.182Z"
}
```

`output` is a single URL string pointing to the restored PNG image.

### 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 an `edm_steps` value above 500.
- **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/photo-realistic-image-restoration-api-cjwbw-supir-v0q";

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: "ede69f6a5ae7d09f769d683347325b08d2f83a93d136ed89747941205e0a71da", 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({
  "image": "https://blog.api.market/wp-content/uploads/2026/10/supir-v0q-input-api-market.png",
  "s_cfg": 7.5,
  "s_churn": 5,
  "s_noise": 1.003,
  "upscale": 1,
  "a_prompt": "Cinematic, High Contrast, highly detailed, taken using a Canon EOS R camera, hyper detailed photo - realistic maximum detail, 32k, Color Grading, ultra HD, extreme meticulous detailing, skin pore detailing, hyper sharpness, perfect without deformations.",
  "min_size": 1024,
  "n_prompt": "painting, oil painting, illustration, drawing, art, sketch, oil painting, cartoon, CG Style, 3D render, unreal engine, blurring, dirty, messy, worst quality, low quality, frames, watermark, signature, jpeg artifacts, deformed, lowres, over-smooth",
  "s_stage1": -1,
  "s_stage2": 1,
  "edm_steps": 50,
  "linear_CFG": false,
  "color_fix_type": "Wavelet",
  "spt_linear_CFG": 1,
  "linear_s_stage2": false,
  "spt_linear_s_stage2": 0
}).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/photo-realistic-image-restoration-api-cjwbw-supir-v0q"


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": "ede69f6a5ae7d09f769d683347325b08d2f83a93d136ed89747941205e0a71da", "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")


output = create_and_wait_for_prediction({
    "image": "https://blog.api.market/wp-content/uploads/2026/10/supir-v0q-input-api-market.png",
    "s_cfg": 7.5,
    "s_churn": 5,
    "s_noise": 1.003,
    "upscale": 1,
    "a_prompt": "Cinematic, High Contrast, highly detailed, taken using a Canon EOS R camera, hyper detailed photo - realistic maximum detail, 32k, Color Grading, ultra HD, extreme meticulous detailing, skin pore detailing, hyper sharpness, perfect without deformations.",
    "min_size": 1024,
    "n_prompt": "painting, oil painting, illustration, drawing, art, sketch, oil painting, cartoon, CG Style, 3D render, unreal engine, blurring, dirty, messy, worst quality, low quality, frames, watermark, signature, jpeg artifacts, deformed, lowres, over-smooth",
    "s_stage1": -1,
    "s_stage2": 1,
    "edm_steps": 50,
    "linear_CFG": False,
    "color_fix_type": "Wavelet",
    "spt_linear_CFG": 1,
    "linear_s_stage2": False,
    "spt_linear_s_stage2": 0
})
print(output)
```

## FAQ
- **Does it enlarge the image?**
  - Only if you ask. `upscale` defaults to 1; set it to 2 or more to enlarge. `min_size` (default 1024) sets the minimum output resolution.
- **Do I need to write a prompt?**
  - No. `a_prompt` and `n_prompt` have defaults tuned for sharp, realistic photos. Change them to steer the result, for example for a specific subject.
- **What does `color_fix_type` do?**
  - It corrects colors after restoration so they stay close to the input. `Wavelet` is the default; `AdaIn` and `None` are the alternatives.
- **What format is the output?**
  - One PNG image, as a URL in `output`.
- **What did the example request cost?**
  - 1 API unit, $0.048, because it took 13.2 seconds of processing.
- **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, 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.


Restore blurry, noisy or low-resolution photos with SUPIR-v0Q, with optional upscaling and prompt-guided detail. $0.048 per API unit, pay as you go.