turbo

Name: Z-Image Turbo
Author: tongyi-mai

High-speed 6B parameter text-to-image generation optimized for cost efficiency and volume. Produces up to 4MP images in an 8-step pipeline suitable for rapid prototyping.

0.005 per megapixel of image

Input

Prompt

STYLE:

Stylized 80s Retro-futurism / Vaporwave digital illustration

Hand-painted texture mixed with smooth neon gradients (like an airbrushed poster)

Lo-fi aesthetic with soft, dreamy bloom and slight film grain overlay

Palette consists of neon pinks, electric blues, deep purples, and vibrant teals

Minimalist composition with bold, clean lines defining the main subject

No text, no letters, no numbers, no logos, no watermark

CANVAS / COMPOSITION:

1:1 square thumbnail

Centered hero object with a strong, recognizable silhouette

Subject occupies ~60% of the frame, surrounded by ample negative space

Background is a clean gradient (e.g., deep twilight purple to hot pink) with a subtle, stylized holographic grid floor vanishing into the distance

High contrast via neon rim lighting against the darker background elements

SUBJECT (use this exact concept):

A floating, stylized cassette tape with tape unraveling into a colorful cosmic river

TEXTURE / LIGHTING / ACTION KEYWORDS:
Texture: matte finish, airbrushed, slight film grain, glowing neon edges
Lighting: harsh backlighting (eclipse style), soft pink/blue ambient glow, dramatic shadows
Action: static, powerful, iconic, floating

RETRO AESTHETIC RULE:

Ensure the image has a tactile, printed poster feel rather than a perfect 3D render.

Use dithering and soft color banding in gradients to simulate vintage digital art.

Do NOT use hyper-realistic textures; keep it illustrative and dreamy.

The prompt to generate an image from.

Image Size

Width

Height

The size of the generated image.

Additional Settings

Customize your input with more control.

Number of Inference Steps

Min: 1 - Max: 8

The number of inference steps to perform.

Seed

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

Number of Images

Min: 1 - Max: 4

The number of images to generate.

Enable Safety Checker

Safety checker can only be disabled on API call

Output Format

The format of the generated image.

Acceleration

The acceleration level to use.

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

Output

{
  "error": "",
  "inferenceTime": 1707,
  "output": [
    "https://media.modelrunner.ai/mxxYaXyiQHxwQXe068Clq.png"
  ],
  "input": {
    "seed": 0,
    "prompt": "STYLE:\n\nStylized 80s Retro-futurism / Vaporwave digital illustration\n\nHand-painted texture mixed with smooth neon gradients (like an airbrushed poster)\n\nLo-fi aesthetic with soft, dreamy bloom and slight film grain overlay\n\nPalette consists of neon pinks, electric blues, deep purples, and vibrant teals\n\nMinimalist composition with bold, clean lines defining the main subject\n\nNo text, no letters, no numbers, no logos, no watermark\n\nCANVAS / COMPOSITION:\n\n1:1 square thumbnail\n\nCentered hero object with a strong, recognizable silhouette\n\nSubject occupies ~60% of the frame, surrounded by ample negative space\n\nBackground is a clean gradient (e.g., deep twilight purple to hot pink) with a subtle, stylized holographic grid floor vanishing into the distance\n\nHigh contrast via neon rim lighting against the darker background elements\n\nSUBJECT (use this exact concept):\n\nA floating, stylized cassette tape with tape unraveling into a colorful cosmic river\n\nTEXTURE / LIGHTING / ACTION KEYWORDS:\nTexture: matte finish, airbrushed, slight film grain, glowing neon edges\nLighting: harsh backlighting (eclipse style), soft pink/blue ambient glow, dramatic shadows\nAction: static, powerful, iconic, floating\n\nRETRO AESTHETIC RULE:\n\nEnsure the image has a tactile, printed poster feel rather than a perfect 3D render.\n\nUse dithering and soft color banding in gradients to simulate vintage digital art.\n\nDo NOT use hyper-realistic textures; keep it illustrative and dreamy.",
    "image_size": "landscape_4_3",
    "num_images": 1,
    "acceleration": "none",
    "output_format": "png",
    "num_inference_steps": 8,
    "enable_safety_checker": false
  },
  "logs": "Generated 1 output(s)"
}

Generated in 1.707 seconds

Logs (1 lines)

Examples

Model Details

Tongyi-MAI's Z-Image Turbo is a streamlined text-to-image generation model engineered for speed and economic scalability. Built on a 6-billion parameter architecture, it prioritizes throughput without sacrificing essential visual coherence. By compressing the diffusion process into a maximum of 8 inference steps, this model significantly reduces generation time compared to standard architectures that typically require 20 to 50 steps.

### Capabilities and Features

This model is specifically tuned for high-volume production environments. Users can generate images with resolutions up to 4 megapixels, supporting various aspect ratios from square to wide landscape. The 8-step pipeline is fully configurable; users can lower the step count to as few as 1 for ultra-fast thumbnail generation or utilize the full 8 steps for final production assets.

**Key benefits include:**

* **Rapid Iteration:** Support for batch sizes up to 4 images per request allows for quick side-by-side comparison of prompts and seeds. * **Flexible Output:** Customize image dimensions via standard presets (e.g., `landscape_4_3`) or specific pixel counts, with support for JPEG, PNG, and WebP formats. * **Cost-Effective Scaling:** The lightweight architecture makes it ideal for applications requiring thousands of assets, such as dynamic content generation or A/B testing visuals.

### When to use this model

Choose Z-Image Turbo when speed and volume are the primary constraints. It excels at rapid prototyping, storyboarding, and generating content variations where cost-per-pixel is a critical metric. While it offers robust prompt adherence, users requiring pixel-perfect photorealism or complex spatial reasoning might prefer larger, slower models.

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: "Cinematic shot of a futuristic cyberpunk street, neon lights, rain on pavement, highly detailed", image_size: "landscape_16_9", num_inference_steps: 8, num_images: 1, enable_safety_checker: true, output_format: "png" } }); ```

tongyi-mai / z-image/turbo

Model Input

Input

Additional Settings

Model Output

Output

Model Example Requests

Examples

Model Details

Model Details