Z-Image Turbo API

Maliyet verimliliği ve hacim için optimize edilmiş, 6B parametreli yüksek hızlı text-to-image üretimi. Hızlı prototiplemeye uygun, 8 adımlı bir pipeline’da 4MP’ye kadar görsel üretir.

görsel megapikseli başına 0.005

OpenAPI

Input

Prompt

STYLE:

Premium 3D render, modern SaaS aesthetic, highly polished

Glassmorphism, frosted acrylic materials, and smooth matte polymers

Soft holographic and iridescent color bleeds inside the glass elements

Clean, minimalist, corporate futuristic tech vibe

Flawless, ultra-smooth surfaces with soft, rounded geometric edges

Octane render, Unreal Engine 5 style, hyper-detailed but visually clean

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

CANVAS / COMPOSITION:

1:1 square thumbnail

Centered hero object with a strong, highly readable silhouette

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

Background is a sleek dark mode gradient (deep charcoal or dark twilight blue) with a subtle, soft-focus ambient glow behind the subject

High contrast between the luminous subject and the dark, clean background

SUBJECT:

A hovering crystalline structure filled with glowing code glyphs

TEXTURE / LIGHTING / ACTION KEYWORDS:
Texture: frosted glass, matte silicone, transparent acrylic, glossy, flawless, smooth
Lighting: subsurface scattering, soft studio lighting, neon rim light, glowing internals, holographic reflections
Action: hovering, levitating, balanced, perfectly still

MATERIAL REFRACTION RULE:

Simulate depth and high-end tech via internal glass refractions and soft glowing cores

Do NOT use heavy grunge, dirt, or gritty textures; everything must remain pristine, sterile, and premium

The prompt to generate an image from.

Image Size

Width

Height

The size of the generated image. Use a preset string (e.g. '4_3_1k') or a custom {width, height} object.

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.

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.

Num Images

Min: 1 - Max: 4

The number of images to generate. Each generated image is billed.

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

Output

{
  "error": "",
  "inferenceTime": 1711,
  "output": [
    "https://media.modelrunner.ai/WbhOaNkhGzMIiPsLHFa5e.png"
  ],
  "input": {
    "seed": 0,
    "prompt": "STYLE:\n\nPremium 3D render, modern SaaS aesthetic, highly polished\n\nGlassmorphism, frosted acrylic materials, and smooth matte polymers\n\nSoft holographic and iridescent color bleeds inside the glass elements\n\nClean, minimalist, corporate futuristic tech vibe\n\nFlawless, ultra-smooth surfaces with soft, rounded geometric edges\n\nOctane render, Unreal Engine 5 style, hyper-detailed but visually clean\n\nNo text, no letters, no numbers, no logos, no watermark, no messy details\n\nCANVAS / COMPOSITION:\n\n1:1 square thumbnail\n\nCentered hero object with a strong, highly readable silhouette\n\nSubject occupies ~60% of the frame, surrounded by generous, clean negative space\n\nBackground is a sleek dark mode gradient (deep charcoal or dark twilight blue) with a subtle, soft-focus ambient glow behind the subject\n\nHigh contrast between the luminous subject and the dark, clean background\n\nSUBJECT:\n\nA hovering crystalline structure filled with glowing code glyphs\n\nTEXTURE / LIGHTING / ACTION KEYWORDS:\nTexture: frosted glass, matte silicone, transparent acrylic, glossy, flawless, smooth\nLighting: subsurface scattering, soft studio lighting, neon rim light, glowing internals, holographic reflections\nAction: hovering, levitating, balanced, perfectly still\n\nMATERIAL REFRACTION RULE:\n\nSimulate depth and high-end tech via internal glass refractions and soft glowing cores\n\nDo NOT use heavy grunge, dirt, or gritty textures; everything must remain pristine, sterile, and premium",
    "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 1.711 seconds

Logs (1 lines)

Örnekler

Z-Image Turbo API

Z-Image Turbo is a text-to-image AI model by tongyi-mai. On ModelRunner it runs through a REST API or via MCP from any AI assistant, at $0.005 per megapixel.

POST https://queue.modelrunner.run/tongyi-mai/z-image/turbo

cURL

# Submit a request to the queue. Input fields go at the top level of the
# body. The optional reserved "metadata" object holds your own flat string
# tags — stored on the request, never sent to the model; filter later with
# GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>.
curl -X POST https://queue.modelrunner.run/tongyi-mai/z-image/turbo \
  -H "Authorization: Key $MRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "seed": 0,
    "prompt": "STYLE:\n\nPremium 3D render, modern SaaS aesthetic, highly polished\n\nGlassmorphism, frosted acrylic materials, and smoo…",
    "image_size": "landscape_4_3",
    "num_images": 1,
    "acceleration": "none",
    "output_format": "png",
    "num_inference_steps": 8,
    "enable_safety_checker": false,
    "metadata": {
      "project": "my-project"
    }
  }'
# → { "request_id": "...", "status_url": "...", "response_url": "..." }

# Poll status_url until "COMPLETED", then fetch the result
curl "https://queue.modelrunner.run/tongyi-mai/z-image/turbo/requests/$REQUEST_ID/status" \
  -H "Authorization: Key $MRUN_API_KEY"
curl "https://queue.modelrunner.run/tongyi-mai/z-image/turbo/requests/$REQUEST_ID" \
  -H "Authorization: Key $MRUN_API_KEY"

JavaScript

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

const result = await modelrunner.subscribe("tongyi-mai/z-image/turbo", {
  input: {
    "seed": 0,
    "prompt": "STYLE:\n\nPremium 3D render, modern SaaS aesthetic, highly polished\n\nGlassmorphism, frosted acrylic materials, and smoo…",
    "image_size": "landscape_4_3",
    "num_images": 1,
    "acceleration": "none",
    "output_format": "png",
    "num_inference_steps": 8,
    "enable_safety_checker": false
  },
});
console.log(result);

Python

import os
import requests

headers = {"Authorization": f"Key {os.environ['MRUN_API_KEY']}"}

submitted = requests.post(
    "https://queue.modelrunner.run/tongyi-mai/z-image/turbo",
    headers=headers,
    json={
      "seed": 0,
      "prompt": "STYLE:\n\nPremium 3D render, modern SaaS aesthetic, highly polished\n\nGlassmorphism, frosted acrylic materials, and smoo…",
      "image_size": "landscape_4_3",
      "num_images": 1,
      "acceleration": "none",
      "output_format": "png",
      "num_inference_steps": 8,
      "enable_safety_checker": false
    },
).json()

# Poll submitted["status_url"] until "COMPLETED", then:
result = requests.get(submitted["response_url"], headers=headers).json()

Input parameters

Input parameters of Z-Image Turbo
Name	Type	Required	Description
prompt	string	yes	The prompt to generate an image from.
image_size	enum	no	The size of the generated image. Use a preset string (e.g. '4_3_1k') or a custom {width, height} object. One of: `1_1_512`, `1_1_1k`, `1_1_2k`, `1_1_4k`, `16_9_512`, `16_9_1k`, `16_9_2k`, `16_9_4k`, `9_16_512`, `9_16_1k`, `9_16_2k`, `9_16_4k`, `4_3_512`, `4_3_1k`, `4_3_2k`, `4_3_4k`, `3_4_512`, `3_4_1k`, `3_4_2k`, `3_4_4k`, `3_2_512`, `3_2_1k`, `3_2_2k`, `3_2_4k`, `2_3_512`, `2_3_1k`, `2_3_2k`, `2_3_4k`, `4_5_512`, `4_5_1k`, `4_5_2k`, `4_5_4k`, `5_4_512`, `5_4_1k`, `5_4_2k`, `5_4_4k`, `4_1_512`, `4_1_1k`, `4_1_2k`, `4_1_4k`, `1_4_512`, `1_4_1k`, `1_4_2k`, `1_4_4k`, `8_1_512`, `8_1_1k`, `8_1_2k`, `8_1_4k`, `1_8_512`, `1_8_1k`, `1_8_2k`, `1_8_4k`, `21_9_512`, `21_9_1k`, `21_9_2k`, `21_9_4k`. Default: "4_3_1k".
num_inference_steps	integer	no	The number of inference steps to perform. Default: 8.
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.
enable_safety_checker	boolean	no	If set to true, the safety checker will be enabled. Default: true.
output_format	enum	no	The format of the generated image. One of: `jpeg`, `png`, `webp`. Default: "png".
acceleration	enum	no	The acceleration level to use. One of: `none`, `regular`, `high`. Default: "regular".
num_images	integer	no	The number of images to generate. Each generated image is billed. Default: 1.

Machine-readable: OpenAPI schema · llms.txt

Use Z-Image Turbo from Claude & Cursor (MCP)

Point Claude Code, Claude Desktop, Cursor, or any MCP client at the ModelRunner MCP server and Z-Image Turbo becomes a tool your assistant can call directly — it authorizes via OAuth (no API key in config) and runs this model with the run_model tool using the endpoint tongyi-mai/z-image/turbo.

MCP client config (Claude Desktop, Cursor)

{
  "mcpServers": {
    "modelrunner": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.modelrunner.run/mcp"]
    }
  }
}

Claude Code

claude mcp add --transport http modelrunner https://mcp.modelrunner.run/mcp

Then ask your assistant, for example: “Run tongyi-mai/z-image/turbo on ModelRunner to generate image”. MCP setup guide.

Model Detayları

Tongyi-MAI’nin Z-Image Turbo modeli, hız ve ekonomik ölçeklenebilirlik için tasarlanmış, sadeleştirilmiş bir text-to-image üretim modelidir. 6 milyar parametreli bir mimari üzerine kuruludur ve temel görsel tutarlılıktan ödün vermeden throughput’u önceler. Diffusion sürecini en fazla 8 inference adımına sıkıştırdığından, genellikle 20 ila 50 adım gerektiren standart mimarilere kıyasla üretim süresini belirgin biçimde kısaltır.

### Yetenekler ve Özellikler

Bu model özellikle yüksek hacimli üretim ortamları için ayarlanmıştır. 4 megapiksele kadar çözünürlükte, kareden geniş yatay formata kadar farklı en-boy oranlarında görseller üretebilirsiniz. 8 adımlı pipeline tümüyle yapılandırılabilir: ultra hızlı thumbnail üretimi için adım sayısını 1’e kadar düşürebilir ya da nihai üretim çıktıları için 8 adımın tamamını kullanabilirsiniz.

**Öne çıkan avantajlar:**

* **Hızlı yineleme:** İstek başına 4 görsele kadar batch desteği, prompt’ları ve seed’leri yan yana hızlıca karşılaştırmanızı sağlar. * **Esnek çıktı:** Görsel boyutlarını standart preset’lerle (ör. `4_3_1k`) ya da belirli piksel sayılarıyla özelleştirin; JPEG, PNG ve WebP desteklenir. * **Uygun maliyetli ölçeklenme:** Hafif mimarisi onu, dinamik içerik üretimi ya da A/B testi görselleri gibi binlerce görsel gerektiren uygulamalar için ideal kılar.

### Bu modeli ne zaman kullanmalı

Hız ve hacim birincil kısıtsa Z-Image Turbo’yu seçin. Hızlı prototipleme, storyboard ve piksel başına maliyetin kritik olduğu içerik varyasyonları üretmede başarılıdır. Prompt uyumu güçlü olsa da, piksel hassasiyetinde fotogerçekçilik ya da karmaşık uzamsal muhakeme gerekiyorsa daha büyük ve daha yavaş modeller tercih edilebilir.

ModelRunner javascript client ile çalıştırmak için aşağıdaki kodu kullanın:

```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: "16_9_1k", num_inference_steps: 8, num_images: 1, enable_safety_checker: true, output_format: "png" } }); ```

Z-Image Turbo API

Model girdisi

Input

Model çıktısı

Output

Örnek istekler

Örnekler

Z-Image Turbo API

cURL

JavaScript

Python

Input parameters

Use Z-Image Turbo from Claude & Cursor (MCP)

MCP client config (Claude Desktop, Cursor)

Claude Code

Model Detayları

Model Detayları