# Z Image | Turbo | Lora

A text-to-image endpoint with LoRA support, powered by Tongyi-MAI’s ultra-fast 6B Z-Image Turbo model for efficient, high-quality image generation.

## API Information

- **Model Slug:** z-image-turbo-lora
- **Branded URL:** https://www.eachlabs.ai/zhipu-ai/z-image/z-image-turbo-lora
- **Provider:** Zhipu AI
- **Category:** Text to Image
- **Output Type:** array
- **Status:** active
- **Version:** 0.0.1
- **Base Cost:** Per-megapixel pricing (round rounding)
- **Estimated Processing Time:** 10 seconds
- **Last Updated:** 2026-04-29
- **Interactive Demo:** https://www.eachlabs.ai/ai-models/z-image-turbo-lora

## Pricing

- **Charge Type:** dynamic
- **Pricing Details:** Per-megapixel pricing (round rounding)

### Pricing Rules

| Condition | Pricing |
| --- | --- |
| output.width < "708" AND output.height < "708" | Minimum 1 MP for small images (< 708x708, round rounding) |
| Rule 2 | Per-megapixel pricing (round rounding) |

## Input Schema

| Parameter | Type | Required | Default | Constraints | Description |
|-----------|------|----------|---------|-------------|-------------|
| prompt | string | Yes | - | - | The prompt to generate an image from. |
| loras | array | No | - | 0–3 | List of LoRA weights to apply |
| image_size | string | No | landscape_4_3 | square_hd,square,portrait_4_3,portrait_16_9,landscape_4_3,landscape_16_9 | The size of the generated image. |
| num_inference_steps | integer | No | 8 | 1–8 | The number of inference steps to perform. |
| seed | integer | No | - | - | The same seed and the same prompt given to the same version of the model will output the same image every time. |
| num_images | integer | No | 1 | 1–4 | The number of images to generate. |
| enable_safety_checker | boolean | No | true | - | If set to true, the safety checker will be enabled. |
| enable_prompt_expansion | boolean | No | false | - | Whether to enable prompt expansion. |
| output_format | string | No | png | jpeg,png,webp | The format of the generated image. |
| acceleration | string | No | none | none,regular,high | The acceleration level to use. |

## Example Request

```bash
curl -X POST https://api.eachlabs.ai/v1/prediction/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "z-image-turbo-lora",
  "input": {
    "prompt": "hyper-realistic, close-up portrait of an Antarctic researcher standing against a backdrop of swirling ice fog and harsh polar winds. Frost clings to the edges of their eyelashes and beard, tiny ice crystals sparkling in the dim blue light. Their weather-beaten skin shows subtle cracks and redness from years of exposure to subzero temperatures. They wear a heavy, fur-lined hood pulled tightly around their face, with frostbitten fabric textures visible in every thread. Reflections of drifting icebergs and a faint aurora glow shimmer in their cold, determined eyes. Shot on a medium-format Hasselblad with high-contrast, gritty winter film grain for a raw, survivalist aesthetic."
  }
}'
```

## Output Schema

Response returned by `GET /v1/prediction/{id}` when the job completes:

```json
{
  "status": "success",
  "predictionID": "string",
  "output": "array",
  "metrics": {
    "predict_time": "number (seconds)"
  }
}
```

## Polling

```bash
curl https://api.eachlabs.ai/v1/prediction/{PREDICTION_ID} \
  -H "Authorization: Bearer YOUR_API_KEY"
```

| Status | Meaning |
|--------|---------|
| `processing` | Still running — poll again |
| `success` | Done — read `output` |
| `error` | Failed — read `message` / `details` |

## Webhook (alternative to polling)

Pass `"webhook_url": "https://your.host/path"` in the create request. Eachlabs POSTs this payload when the job ends:

```json
{
  "exec_id": "prediction-uuid",
  "status": "succeeded",
  "output": "https://...",
  "error": ""
}
```

`status` is `"succeeded"` or `"failed"`. `exec_id` equals the `predictionID` from create. Return 2xx within 30 seconds.

## Errors

Error body: `{ "status": "error", "message": "...", "details": "..." }`

| Code | Meaning |
|------|---------|
| `400` | Invalid input |
| `401` | Missing / invalid `Authorization` bearer token |
| `404` | Unknown model or prediction id |
| `429` | Rate limit — 100 creates / min, 10 concurrent per key |
| `5xx` | Retry with backoff |

## Overview

**z-image-turbo-lora — Text-to-Image AI Model**

z-image-turbo-lora, developed by Zhipu AI as part of the z-image family, is a text-to-image generation model powered by Tongyi-MAI's ultra-fast 6 billion parameter architecture. It solves the core problem facing developers and creators: generating photorealistic images quickly without sacrificing quality or requiring massive computational resources. Unlike standard text-to-image models that demand high VRAM and extended processing times, z-image-turbo-lora delivers high-fidelity outputs in seconds on consumer-grade hardware, making it ideal for building responsive AI image generation APIs and applications.

The model's primary strength lies in its efficiency-to-quality ratio. It generates 1024×1024 photorealistic images in approximately 9 seconds on an RTX 4080, with support for external LoRA modules that enable custom style adaptation without retraining. This combination of speed, quality, and extensibility positions z-image-turbo-lora as a practical choice for developers building production text-to-image AI systems that need to balance latency, cost, and visual fidelity.

## Usage Notes

- API Base URL: `https://api.eachlabs.ai/v1`
- Authentication: send `Authorization: Bearer YOUR_API_KEY`. Generate a key from the Eachlabs dashboard at https://www.eachlabs.ai/dashboard/api-keys.
- File-typed parameters (`*_url`, `image_url`, `video_url`, `audio_url`, etc.) accept publicly-reachable HTTPS URLs only. Upload your asset first (GCS / S3 / your CDN) and pass the resulting URL. Data-URIs and localhost URLs are rejected.
- For structured parameters (arrays / objects) send real JSON values, not stringified payloads.
- Monetary values are reported in USD; per-token / per-megapixel rates may be billed in micro-cents internally.
- Prefer `webhook_url` over polling for long-running predictions — see the Webhook Callback section.