Name: Tongyi-MAI Z-Image Turbo Text-to-Image Generator API
Brand: magicapi
SKU: api-magicapi-ultra-fast-text-to-image-generator-prunaai-z-image-turbo
Availability: InStock
Author: API Market

Shashank Agarwal

# ⚡ Text-to-Image Generator API (Tongyi-MAI Z-Image Turbo)

## Overview
Z-Image Turbo is a text-to-image model with 6B parameters developed by Tongyi-MAI. Its developer describes it as super fast. This endpoint runs it in a build by Pruna AI.

Send a text `prompt` with the `width` and `height` you want (64 to 2048 pixels each). Each request returns the URL of one image.

## Features
- **Fast generation**: 8 inference steps by default (`num_inference_steps`, 1 to 50).
- **Any size**: set `width` and `height` from 64 to 2048 pixels each (default 1024 x 1024).
- **Guidance-free**: `guidance_scale` defaults to 0, the value recommended for this Turbo model; negative prompts are not used.
- **Output formats**: `jpg` (default), `png` or `webp`, with `output_quality` from 0 to 100 for jpg and webp.
- **Optional safety filter**: `enable_safety_filter` checks the prompt against the content policy. When it is off, only the always-on categories are blocked.
- **Seed**: set `seed` to reproduce a result.

## Use Cases
1. **Draft visuals quickly**: try many prompt variations in seconds for concepts and storyboards.
2. **Product and lifestyle shots**: studio-style images for listings, mockups and catalogs.
3. **Images at an exact size**: banners, thumbnails and social images generated at the pixel size they will be shown.
4. **Input for other models**: people, objects and scenes to feed into editing, try-on or video models.

## Visual Examples
### Example Input
Prompt: "Full-length studio photo of a young woman standing straight and facing the camera, arms relaxed at her sides, wearing a plain white t-shirt and light blue jeans, plain light grey background, soft even lighting", at 576 x 768 pixels with 8 inference steps.

```json
{
  "prompt": "Full-length studio photo of a young woman standing straight and facing the camera, arms relaxed at her sides, wearing a plain white t-shirt and light blue jeans, plain light grey background, soft even lighting",
  "width": 576,
  "height": 768,
  "output_format": "jpg",
  "guidance_scale": 0,
  "output_quality": 80,
  "num_inference_steps": 8
}
```

### Example Output
![Z-Image Turbo output: full-length studio photo of a young woman in a white t-shirt and light blue jeans against a light grey background](https://blog.api.market/wp-content/uploads/2026/11/z-image-turbo-output-zfy68yms-api-market.jpg)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | Yes |  |  | Text prompt for image generation. |
| `height` | integer | No | `1024` | 64 to 2048 | Height of the generated image in pixels. |
| `width` | integer | No | `1024` | 64 to 2048 | Width of the generated image in pixels. |
| `num_inference_steps` | integer | No | `8` | 1 to 50 | Number of inference steps. |
| `guidance_scale` | number | No | `0` | 0 to 20 | Guidance scale. Keep it at 0 for this Turbo model. |
| `seed` | integer | No | none |  | Random seed. Set it for reproducible generation. |
| `go_fast` | boolean | No | `false` |  | Unused compatibility field. This endpoint always runs the bf16 model. |
| `output_format` | string | No | `jpg` | `png`, `jpg`, `webp` | Format of the output image. |
| `output_quality` | integer | No | `80` | 0 to 100 | Quality when saving the output image, from 0 to 100 (100 is best). Not used for png. |
| `num_outputs` | integer | No | `1` | 1 to 4 | Number of output images. Only 1 is currently supported. |
| `safety_tolerance` | integer | No | `2` | 1 to 6 | Unused compatibility field. Use enable_safety_filter instead. |
| `model_version` | string | No | none |  | Unused compatibility field. This endpoint always runs Z-Image Turbo. |
| `negative_prompt` | string | No | none |  | Unused compatibility field. This model is guidance-free (guidance_scale 0) and does not use negative prompts. |
| `aspect_ratio` | string | No | none |  | Unused compatibility field. Set width and height instead. |
| `megapixels` | string | No | none |  | Unused compatibility field. Set width and height instead. |
| `disable_safety_checker` | boolean | No | `true` |  | Unused compatibility field. Use enable_safety_filter instead. |
| `enable_safety_checker` | boolean | No | `false` |  | Unused compatibility field. Use enable_safety_filter instead. |
| `enable_safety_filter` | boolean | No | `false` |  | Check the prompt against the content policy. When off, only the always-on categories (sexual content involving minors, self-harm intent or instructions) are still blocked. |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.001 per API unit.
- **Metered billing**: each image is billed at the price of its resolution bucket, times the billed count the model reports for it (1 for a 2048 x 2048 image, 1.25 for the 576 x 768 example). The total is rounded up to a whole API unit per request.
- **Price per output image by resolution bucket** (the model reports the bucket from the output size):

| Resolution bucket | Price per output image (billed count 1) |
| --- | --- |
| `0.5` | $0.00295 |
| `1` | $0.00589 |
| `2` | $0.0118 |
| `3` | $0.0177 |
| `4` | $0.0236 |

- No request field selects the bucket; it follows `width` x `height`. If the model reports no bucket, the `4` price applies. If it reports no billed count, each request is billed at least 1.25 per image.
- **Example cost**: the 576 x 768 example image fell in the `0.5` bucket and the model reported a billed count of 1.25: 1.25 x $0.00295 = $0.0037, rounded up to 4 units ($0.004).
- **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/ultra-fast-text-to-image-generator-prunaai-z-image-turbo`

### 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/ultra-fast-text-to-image-generator-prunaai-z-image-turbo/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "prunaai/z-image-turbo",
    "input": {
      "prompt": "Full-length studio photo of a young woman standing straight and facing the camera, arms relaxed at her sides, wearing a plain white t-shirt and light blue jeans, plain light grey background, soft even lighting",
      "width": 576,
      "height": 768,
      "output_format": "jpg",
      "guidance_scale": 0,
      "output_quality": 80,
      "num_inference_steps": 8
    }
  }'
```

Response (201 Created):
```json
{
  "id": "zfy68ymsz5rmr0d1n17sp99524",
  "version": "prunaai/z-image-turbo",
  "status": "starting",
  "created_at": "2026-11-04T20:45:51.481Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/ultra-fast-text-to-image-generator-prunaai-z-image-turbo/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "zfy68ymsz5rmr0d1n17sp99524",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/11/z-image-turbo-output-zfy68yms-api-market.jpg",
  "error": null,
  "created_at": "2026-11-04T20:45:51.481Z",
  "started_at": "2026-11-04T20:45:51.496Z",
  "completed_at": "2026-11-04T20:45:53.364Z"
}
```

`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 `width` above 2048.
- **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/ultra-fast-text-to-image-generator-prunaai-z-image-turbo";

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: "prunaai/z-image-turbo", 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({
  "prompt": "Full-length studio photo of a young woman standing straight and facing the camera, arms relaxed at her sides, wearing a plain white t-shirt and light blue jeans, plain light grey background, soft even lighting",
  "width": 576,
  "height": 768,
  "output_format": "jpg",
  "guidance_scale": 0,
  "output_quality": 80,
  "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/ultra-fast-text-to-image-generator-prunaai-z-image-turbo"


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": "prunaai/z-image-turbo", "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({
    "prompt": "Full-length studio photo of a young woman standing straight and facing the camera, arms relaxed at her sides, wearing a plain white t-shirt and light blue jeans, plain light grey background, soft even lighting",
    "width": 576,
    "height": 768,
    "output_format": "jpg",
    "guidance_scale": 0,
    "output_quality": 80,
    "num_inference_steps": 8
}))
```

## FAQ
- **How many images does one request return?**
  - One. `num_outputs` accepts 1 to 4, but only 1 is currently supported.
- **Should I change `guidance_scale`?**
  - No. Keep it at 0, the default. The Turbo model is guidance-free and ignores `negative_prompt`.
- **Why are some fields marked unused?**
  - `aspect_ratio`, `megapixels`, `model_version`, `go_fast`, `negative_prompt` and the safety checker fields exist for compatibility with other clients and have no effect. Set the size with `width` and `height`.
- **How is the price of an image decided?**
  - The model reports a resolution bucket and a billed count for each image (1 for a 2048 x 2048 image, 1.25 for the 576 x 768 example). You pay that count times the bucket's price, rounded up to a whole $0.001 unit per request. The 576 x 768 example cost $0.004.
- **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.


Generate images from text with Z-Image Turbo, a fast 6B-parameter model from Tongyi-MAI, at 64 to 2048 pixels per side. From $0.00295 per image, pay as you go.