Logo
LoginGet Started
Profile

tongyi-mai / z-image/turbo(deprecated)

Generate high-volume imagery rapidly with this 6B-parameter text-to-image model. Optimized for speed and cost-efficiency, it supports up to 4-megapixel resolutions and batch generation, making it ideal for rapid prototyping and variation testing.

0.005 per megapixel of image

Model Input

Input

The prompt to generate an image from.

The size of the generated image.

Min: 1 - Max: 4

The number of images to generate.

Additional Settings

Customize your input with more control.

Min: 1 - Max: 8

The number of inference steps to perform.

The same seed and the same prompt given to the same version of the model will output the same image every time.

Safety checker can only be disabled on API call

The acceleration level to use.

The format of the generated image.

Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.

If `True`, the media will be returned as a data URI and the output data won't be available in the request history.

You need to be logged in to run this model and view results.
Log in

Model Output

Output

Fill in the input form and click submit to see the output
Logs (1 lines)

Model Details

Model Details

# Z-Image Turbo [text-to-image]

Z-Image Turbo is a high-speed text-to-image generation model designed specifically for workflows where throughput, cost-efficiency, and speed are paramount. Built on a lean 6B-parameter architecture, it significantly reduces generation time by compressing the inference pipeline to a maximum of 8 steps, contrasting sharply with standard diffusion models that often require 20-50 steps.

### Key Capabilities

* **Speed-Optimized Inference:** The model allows users to configure inference steps between 1 and 8. Use lower steps for instant thumbnails and rapid iteration, or the full 8 steps for higher-quality final assets. * **High Resolution & Flexibility:** despite its speed, the model supports resolutions up to 4 megapixels. Users can select from various aspect ratios via the UI, including square, portrait, and landscape formats (e.g., `landscape_16_9`, `portrait_4_3`). * **Batch Generation:** To facilitate rapid variation testing, the model supports generating up to 4 images in a single API call. * **Prompt Enhancement:** The model features an optional "Prompt Expansion" setting. When enabled, it automatically enriches brief prompts with descriptive details to produce more robust and visually interesting results without requiring complex prompt engineering.

### Best Use Cases

This model is ideal for high-volume applications such as rapid prototyping, storyboard creation, and content variation testing. While it trades some of the extreme photorealism found in larger, slower models for raw speed, it maintains excellent prompt adherence and is perfect for production environments requiring thousands of assets.

To run via ModelRunner javascript client, use the following code:

```javascript import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe('tongyi-mai/z-image/turbo', { input: { "prompt": "A futuristic city skyline at sunset, cyberpunk style, neon lights", "image_size": "landscape_16_9", "num_images": 1, "num_inference_steps": 8, "enable_prompt_expansion": true, "output_format": "jpeg", "enable_safety_checker": true } }); ```