Name: Krea 3 Image Editing and Generation API
Brand: magicapi
SKU: api-magicapi-image-editing-generation-krea-krea-3
Availability: InStock
Author: API Market

Shashank Agarwal

# 🖌️ Image Editing and Generation API (Krea 3)

## Overview
Krea 3 is an image model from Krea. Its developer describes it as a fast image editing model that follows plain-language instructions; it also generates images from text.

Send up to 8 `images` with a `prompt` that says what should change and what should stay the same, or leave `images` empty to generate an image from the prompt alone. Each request returns the URL of one image.

## Features
- **Instruction-based editing**: describe the edit in `prompt`; for several images, refer to each by what it shows.
- **Text to image**: leave `images` empty (the default) to generate from the prompt.
- **Up to 8 input images** to edit or combine.
- **Thinking mode**: `fast` (default) for most requests, `high` to plan complex or multi-step edits more carefully (slower, higher price).
- **Resolution**: `1k` or `2k` (default).
- **Aspect ratios**: `auto` (default, framed from the inputs and prompt), `1:1`, `4:5`, `3:4`, `2:3`, `9:16`, `5:4`, `4:3`, `3:2`, `16:9`, `21:9`.
- **Regional edit (experimental)**: `regional_edit` changes only the parts that need to change. Needs at least one input image.
- **Negative prompt and seed**: `negative_prompt` lists what to avoid; `seed` does not guarantee identical results because the prompt is expanded before generation.

## Use Cases
1. **Photo restyling**: turn a photo into a painting, sketch or another look while keeping its subject.
2. **Product image edits**: change backgrounds, colors or props in catalog photos.
3. **Combining images**: put an object from one image into the scene of another.
4. **New images from text**: concept art and marketing visuals at 1k or 2k.

## Visual Examples
### Example Input
Our bird photo with the prompt "Turn this photo into a watercolor painting on textured paper. Keep the bird, the branch and the composition unchanged.", at `1k` with `thinking_mode` `fast` and `aspect_ratio` `auto`.

![Input: a colorful bird perched on a branch in front of a field of pink flowers](https://blog.api.market/wp-content/uploads/2026/10/supir-input-api-market.png)

```json
{
  "images": [
    "https://blog.api.market/wp-content/uploads/2026/10/supir-input-api-market.png"
  ],
  "prompt": "Turn this photo into a watercolor painting on textured paper. Keep the bird, the branch and the composition unchanged.",
  "resolution": "1k",
  "aspect_ratio": "auto",
  "regional_edit": false,
  "thinking_mode": "fast"
}
```

### Example Output
![Krea 3 output: the same bird and branch as a watercolor painting on textured paper](https://blog.api.market/wp-content/uploads/2026/11/krea-3-output-ryy9p4cv-api-market.jpg)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | Yes |  |  | Instruction describing the image to generate, or the edit to make. For edits, say what should change and what should stay the same. |
| `images` | array of URLs | No | `[]` |  | Up to 8 input images (URLs) to edit or combine. Leave empty for text-to-image generation. When using several images, refer to each by what it shows in the prompt. |
| `thinking_mode` | string | No | `fast` | `fast`, `high` | How much planning happens before generation. `fast` is quicker and works for most requests. `high` plans complex or multi-step edits more carefully, but is slower and costs more. |
| `resolution` | string | No | `2k` | `1k`, `2k` | Output resolution. |
| `aspect_ratio` | string | No | `auto` | `auto`, `1:1`, `4:5`, `3:4`, `2:3`, `9:16`, `5:4`, `4:3`, `3:2`, `16:9`, `21:9` | Aspect ratio of the output. `auto` picks the framing from your input images and prompt. |
| `regional_edit` | boolean | No | `false` |  | Experimental. Change only the parts of the image that need to change and keep the rest unchanged. Needs at least one input image. Quality may be lower and generation slower than a regular edit. |
| `negative_prompt` | string | No | none |  | What to avoid in the output. |
| `seed` | integer | No | none |  | Random seed. Reusing a seed does not guarantee identical results, because the prompt is expanded before generation. |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.001 per API unit.
- **Metered billing**: a request is billed for what it uses, summed over each priced item (output images and input image megapixels), then rounded up to a whole API unit.
- **Price by setting**:

| `thinking_mode` | `resolution` | Per output image | Per input image megapixel |
| --- | --- | --- | --- |
| `fast` | `1k` | $0.0353 | $0.0236 |
| `high` | `1k` | $0.0589 | $0.0236 |
| `fast` | `2k` | $0.0706 | $0.0236 |
| `high` | `2k` | $0.0942 | $0.0236 |

- When `thinking_mode` and `resolution` are left out, `fast` at `2k` applies: a text-to-image request with no input images costs 71 units ($0.071).
- Input image megapixels are the total size of the `images` you send, as the model measures it (an image that cannot be read counts as 12 megapixels). Text-to-image requests have none.
- **Example cost**: the example returned 1 image at `fast` and `1k` from a 0.262 megapixel input image: 1 x $0.0353 + 0.262 x $0.0236 = $0.0415, rounded up to 42 units ($0.042).
- **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-editing-generation-krea-krea-3`

### 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-editing-generation-krea-krea-3/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "krea/krea-3",
    "input": {
      "images": [
        "https://blog.api.market/wp-content/uploads/2026/10/supir-input-api-market.png"
      ],
      "prompt": "Turn this photo into a watercolor painting on textured paper. Keep the bird, the branch and the composition unchanged.",
      "resolution": "1k",
      "aspect_ratio": "auto",
      "regional_edit": false,
      "thinking_mode": "fast"
    }
  }'
```

Response (201 Created):
```json
{
  "id": "ryy9p4cv0xrmy0d1n17szeqham",
  "version": "krea/krea-3",
  "status": "starting",
  "created_at": "2026-11-04T20:45:51.751Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/image-editing-generation-krea-krea-3/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "ryy9p4cv0xrmy0d1n17szeqham",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/11/krea-3-output-ryy9p4cv-api-market.jpg",
  "error": null,
  "created_at": "2026-11-04T20:45:51.751Z",
  "started_at": "2026-11-04T20:45:51.765Z",
  "completed_at": "2026-11-04T20:45:58.600Z"
}
```

`output` is a single URL string pointing to the generated 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 `prompt` or a `resolution` that is not `1k` or `2k`.
- **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
Requests are rate limited to keep usage fair. If you receive a 429 response, wait and retry with exponential backoff.

## 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-editing-generation-krea-krea-3";

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: "krea/krea-3", 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({
  "images": [
    "https://blog.api.market/wp-content/uploads/2026/10/supir-input-api-market.png"
  ],
  "prompt": "Turn this photo into a watercolor painting on textured paper. Keep the bird, the branch and the composition unchanged.",
  "resolution": "1k",
  "aspect_ratio": "auto",
  "regional_edit": false,
  "thinking_mode": "fast"
}).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-editing-generation-krea-krea-3"


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": "krea/krea-3", "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({
    "images": [
        "https://blog.api.market/wp-content/uploads/2026/10/supir-input-api-market.png"
    ],
    "prompt": "Turn this photo into a watercolor painting on textured paper. Keep the bird, the branch and the composition unchanged.",
    "resolution": "1k",
    "aspect_ratio": "auto",
    "regional_edit": False,
    "thinking_mode": "fast"
}))
```

## FAQ
- **Can I generate an image without an input image?**
  - Yes. Leave `images` empty (the default) and describe the image in `prompt`.
- **How many images can I send?**
  - Up to 8 in `images`. Refer to each by what it shows in the prompt, for example "the lamp from the second image".
- **When should I use `thinking_mode` `high`?**
  - For complex or multi-step edits that need more planning. It is slower and costs more per output image ($0.0589 at 1k, $0.0942 at 2k).
- **Why does the same seed give a different result?**
  - The prompt is expanded before generation, so reusing a `seed` does not guarantee identical results.
- **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`.
- **What if my prediction fails?**
  - Failed predictions are not billed. Read the `error` field, adjust the input and try again.

## Contact Us
If you have any questions, reach out to us through the chat on the bottom right of the page.


Edit and combine up to 8 images with plain-language instructions, or generate images from text, with Krea 3 at 1k or 2k. From $0.0353 per output image, pay as you go.