---
title: "z-image-turbo-lora"
url: "https://docs.api.pruna.ai/guides/models/z-image-turbo-lora"
image: "https://docs.api.pruna.ai/_og/d/c_Ocean.takumi,title_z-image-turbo-lora,props_eyJ0aGVtZSI6eyJtb2RlIjoibGlnaHQiLCJjb2xvcnMiOnsicHJpbWFyeSI6IiM3YzNhZWQifX19,p_Ii9ndWlkZXMvbW9kZWxzL3otaW1hZ2UtdHVyYm8tbG9yYSI,s_jEYxyiR39U9JRcS6.png"
---

# Z Image Turbo Lora

Z-image with LoRA support for unique styles and personalised generations

## [Overview](#overview)

Z-Image-Turbo-LoRA extends the fast generation capabilities of Z-Image-Turbo with support for LoRA (Low-Rank Adaptation) weights. Load custom LoRA models from HuggingFace or other sources to apply unique artistic styles, characters, or concepts to your generations. Supports multiple LoRAs simultaneously with adjustable scales.

**Rate Limit:** 150 requests per minute

**Category:** Image Generation

**Price:** 0.008$ / image

---

## [Quickstart](#quickstart)

### [Basic Image Generation (Asynchronous)](#basic-image-generation-asynchronous)

```bash
curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Model: z-image-turbo-lora' \
  -d '{
    "input": {
      "prompt": "Portrait of a warrior in epic fantasy armor"
    }
  }'
```

### [Basic Image Generation (Synchronous)](#basic-image-generation-synchronous)

```bash
curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Model: z-image-turbo-lora' \
  -H 'Try-Sync: true' \
  -d '{
    "input": {
      "prompt": "Portrait of a warrior in epic fantasy armor"
    }
  }'
```

### [Single LoRA Customization](#single-lora-customization)

```bash
curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Model: z-image-turbo-lora' \
  -d '{
    "input": {
      "prompt": "Anime style portrait of a young woman",
      "lora_weights": ["https://huggingface.co/your-user/your-lora/resolve/main/lora.safetensors"],
      "lora_scales": [0.8],
      "seed": 42
    }
  }'
```

### [Multiple LoRA Combination](#multiple-lora-combination)

```bash
curl -X POST 'https://api.pruna.ai/v1/predictions' \
  -H 'Content-Type: application/json' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Model: z-image-turbo-lora' \
  -d '{
    "input": {
      "prompt": "Fantasy landscape with watercolor painting style",
      "lora_weights": [
        "https://huggingface.co/user1/style-lora/resolve/main/lora1.safetensors",
        "https://huggingface.co/user2/concept-lora/resolve/main/lora2.safetensors"
      ],
      "lora_scales": [0.7, 0.5],
      "width": 1440,
      "height": 768,
      "output_format": "png"
    }
  }'
```

---

## [Parameters](#parameters)

### [Required Parameters](#required-parameters)

| Parameter | Type   | Description                      |
| :-------- | :----- | :------------------------------- |
| prompt    | string | Text prompt for image generation |

### [Optional Parameters](#optional-parameters)

| Parameter           | Type             | Default    | Description                                                                      |
| :------------------ | :--------------- | :--------- | :------------------------------------------------------------------------------- |
| height              | integer          | 1024       | Height of the generated image (64-1440)                                          |
| width               | integer          | 1024       | Width of the generated image (64-1440)                                           |
| num_inference_steps | integer          | 8          | Number of inference steps. This actually results in (num_inference_steps - 1) DiT forwards (1-50) |
| guidance_scale      | float            | 0.0        | Guidance scale. Should be 0 for Turbo models (0.0-20.0)                          |
| seed                | integer          | (optional) | Random seed. Set for reproducible generation                                     |
| lora_weights        | array of strings | (optional) | LoRA weights as an array of URLs. Supports .safetensors URLs, tar files, and zip files from HuggingFace or other sources (e.g., 'https://huggingface.co/user/model/resolve/main/lora.safetensors') |
| lora_scales         | array of floats  | (optional) | Scales for each LoRA as an array of floats. Must match the number of lora_weights. Defaults to 1.0 for each if not provided |
| output_format       | string           | "jpg"      | Format of the output images. Options: "png", "jpg", "webp"                       |
| output_quality      | integer          | 80         | Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs |

This is your starter page content.