Name: Kandinsky 2.2 API: Multilingual Text to Image Generator
Brand: magicapi
SKU: api-magicapi-multilingual-text-to-image-generator-ai-forever-kandinsky-2-2
Availability: InStock
Author: API Market

Shashank Agarwal

# 🎨 Multilingual Text-to-Image Generator API (ai-forever Kandinsky 2.2)

## Overview
Kandinsky 2.2 is a latent diffusion text-to-image model published by ai-forever. Its developer describes it as a multilingual text-to-image model.

Send a text `prompt` and the API returns a list of image URLs. You choose the width and height from a fixed list of sizes (384 to 2048 pixels), the number of images (1 to 4), the output format and the number of denoising steps.

## Features
- **Text to image**: generate images from a written `prompt`.
- **Negative prompt**: describe what you do not want to see with `negative_prompt`.
- **Fixed sizes**: `width` and `height` each accept 384, 512, 576, 640, 704, 768, 960, 1024, 1152, 1280, 1536, 1792 or 2048 (default 512).
- **Up to 4 images per request**: set `num_outputs` from 1 to 4.
- **Step control**: `num_inference_steps` (default 75) and `num_inference_steps_prior` (default 25), each from 1 to 500.
- **Output formats**: `webp` (default), `jpeg` or `png`.
- **Seed**: set `seed` to reproduce a result; leave it empty for a random seed.

## Use Cases
1. **Illustrations and concept art**: turn short scene descriptions into images.
2. **Prompts in several languages**: write prompts in the language of your users.
3. **Design variations**: request up to 4 images at once and pick the best one.
4. **Social and blog visuals**: square or wide images in a fixed set of sizes.

## Visual Examples
### Example Input
Prompt: "A moss covered astronaut with a black background"

Settings: `width` 1024, `height` 1024, `num_outputs` 1, `num_inference_steps` 75. All other fields were left at their defaults.

### Example Output
![Kandinsky 2.2 output generated from the moss covered astronaut example prompt](https://blog.api.market/wp-content/uploads/2026/10/kandinsky-2-2-output-9tfn87qp-api-market.webp)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | No | `A moss covered astronaut with a black background` |  | Input prompt |
| `negative_prompt` | string | No | none |  | Things you do not want to see in the output |
| `width` | integer | No | `512` | `384`, `512`, `576`, `640`, `704`, `768`, `960`, `1024`, `1152`, `1280`, `1536`, `1792`, `2048` | Width of the output image. Lower it if the request runs out of memory |
| `height` | integer | No | `512` | `384`, `512`, `576`, `640`, `704`, `768`, `960`, `1024`, `1152`, `1280`, `1536`, `1792`, `2048` | Height of the output image. Lower it if the request runs out of memory |
| `num_inference_steps` | integer | No | `75` | 1 to 500 | Number of denoising steps |
| `num_inference_steps_prior` | integer | No | `25` | 1 to 500 | Number of denoising steps for the prior |
| `num_outputs` | integer | No | `1` | 1 to 4 | Number of images to output |
| `seed` | integer | No | none |  | Random seed. Leave blank to randomize the seed |
| `output_format` | string | No | `webp` | `webp`, `jpeg`, `png` | Output image format |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.028 per API unit.
- **1 API unit = 1 successful request with up to 18 seconds of processing.** A request that takes longer is billed one more unit for each started block of 18 seconds (for example, 20 seconds of processing is 2 units, $0.056).
- Larger sizes, more steps and more images per request take longer to process and can use more than one unit.
- **Example cost**: the example request above took 7.3 seconds of processing, so it used 1 API unit and cost $0.028.
- **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/multilingual-text-to-image-generator-ai-forever-kandinsky-2-2`

### 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/multilingual-text-to-image-generator-ai-forever-kandinsky-2-2/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "ad9d7879fbffa2874e1d909d1d37d9bc682889cc65b31f7bb00d2362619f194a",
    "input": {
      "width": 1024,
      "height": 1024,
      "prompt": "A moss covered astronaut with a black background",
      "num_outputs": 1,
      "num_inference_steps": 75
    }
  }'
```

Response (201 Created):
```json
{
  "id": "9tfn87qpgnrna0d1cchbagj4pw",
  "version": "ad9d7879fbffa2874e1d909d1d37d9bc682889cc65b31f7bb00d2362619f194a",
  "status": "starting",
  "created_at": "2026-10-22T10:23:18.318Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/multilingual-text-to-image-generator-ai-forever-kandinsky-2-2/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "9tfn87qpgnrna0d1cchbagj4pw",
  "status": "succeeded",
  "output": [
    "https://blog.api.market/wp-content/uploads/2026/10/kandinsky-2-2-output-9tfn87qp-api-market.webp"
  ],
  "error": null,
  "created_at": "2026-10-22T10:23:18.318Z",
  "started_at": "2026-10-22T10:25:09.373Z",
  "completed_at": "2026-10-22T10:25:16.719Z"
}
```

`output` is a list of image URLs, one per image (`num_outputs`).

### 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 `width` that is not in the list or `num_outputs` above 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.
- If a large size fails with a memory error, lower `width` and `height` and try again.
- 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/multilingual-text-to-image-generator-ai-forever-kandinsky-2-2";

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: "ad9d7879fbffa2874e1d909d1d37d9bc682889cc65b31f7bb00d2362619f194a", 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({
  width: 1024,
  height: 1024,
  prompt: "A moss covered astronaut with a black background",
  num_outputs: 1,
  num_inference_steps: 75
}).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/multilingual-text-to-image-generator-ai-forever-kandinsky-2-2"


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": "ad9d7879fbffa2874e1d909d1d37d9bc682889cc65b31f7bb00d2362619f194a", "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({
    "width": 1024,
    "height": 1024,
    "prompt": "A moss covered astronaut with a black background",
    "num_outputs": 1,
    "num_inference_steps": 75
}))
```

## FAQ
- **What sizes can I request?**
  - `width` and `height` each take one value from 384, 512, 576, 640, 704, 768, 960, 1024, 1152, 1280, 1536, 1792 and 2048. The default is 512 x 512.
- **How many images does one request return?**
  - `num_outputs` images, from 1 to 4. `output` is always a list of URLs.
- **How is a request billed?**
  - $0.028 per request for up to 18 seconds of processing, plus one more unit for each further started block of 18 seconds. The example request (1024 x 1024, 75 steps) took 7.3 seconds and cost $0.028.
- **What happens if I leave out the prompt?**
  - The model uses its default prompt, "A moss covered astronaut with a black background". Send your own `prompt` for any real use.
- **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. You can also email our support team at hello@magicapi.com.


Generate images from text prompts with Kandinsky 2.2, a multilingual latent diffusion model, in sizes from 384 to 2048 pixels. Pay as you go, no monthly fee.