Skip to main content
Endpoint: POST https://fal.run/fal-ai/z-image/turbo Endpoint ID: fal-ai/z-image/turbo

Try it in the Playground

Run this model interactively with your own prompts.

​
Quick Start

import fal_client

def on_queue_update(update):
    if isinstance(update, fal_client.InProgress):
        for log in update.logs:
           print(log["message"])

result = fal_client.subscribe(
    "fal-ai/z-image/turbo",
    arguments={
        "prompt": "A hyper-realistic, close-up portrait of a tribal elder from the Omo Valley, painted with intricate white chalk patterns and adorned with a headdress made of dried flowers, seed pods, and rusted bottle caps. The focus is razor-sharp on the texture of the skin, showing every pore, wrinkle, and scar that tells a story of survival. The background is a blurred, smoky hut interior, with the warm glow of a cooking fire reflecting in the subject's dark, soulful eyes. Shot on a Leica M6 with Kodak Portra 400 film grain aesthetic."
    },
    with_logs=True,
    on_queue_update=on_queue_update,
)
print(result)
Tongyi-MAI’s Z-Image Turbo delivers 6B-parameter text-to-image generation at $0.005 per megapixel. Trading model size for raw speed through an 8-step inference pipeline, it optimizes for high-volume applications where cost efficiency and throughput matter more than maximum detail fidelity. Built for: Rapid prototyping | Content variation testing | High-volume asset generation

​
Speed-First Architecture for Volume Workflows

Z-Image Turbo compresses inference to 8 steps maximum (configurable down to 1), contrasting with standard diffusion models that typically require 20-50 steps for comparable output quality. The 6B parameter count keeps memory footprint lean while maintaining prompt adherence. What this means for you:
  • Flexible resolution up to 4 megapixels: Generate images from portrait to landscape orientations without resolution caps limiting your workflow
  • Batch generation up to 4 images: Test multiple variations in a single API call for faster iteration cycles
  • Configurable inference steps (1-8): Trade quality for speed based on your specific use case, use 1-step for thumbnails, 8-step for final assets
  • Optional prompt expansion: Automatically enhance brief prompts with descriptive detail when you need richer outputs (adds $0.0025 per request)

​
Performance That Scales

Z-Image Turbo optimizes for cost-per-image economics in production environments where you’re generating hundreds or thousands of assets.
MetricResultContext
Cost per Megapixel$0.005200 megapixels per $1.00 on fal
Inference Steps1-8 configurableDefault 8 steps balances speed/quality
Max ResolutionUp to 4MPSupports standard aspect ratios from square to ultrawide
Batch Size1-4 imagesGenerate variations in single request

​
Technical Specifications

SpecDetails
ArchitectureZ-Image Turbo (6B parameters)
Input FormatsText prompts with optional seed control
Output FormatsJPEG, PNG, WebP
Max Resolution4 megapixels (configurable aspect ratios)
TrainingZ-Image Trainer available for LoRA fine-tuning
LicenseCommercial use permitted
API Documentation | Quickstart Guide

​
How It Stacks Up

AuraFlow – Z-Image Turbo prioritizes inference speed and cost efficiency through its 8-step maximum pipeline, making it ideal for high-volume generation workflows where throughput matters. AuraFlow emphasizes prompt interpretation depth for complex creative briefs requiring nuanced understanding. FLUX.2 [dev] – Z-Image Turbo trades parameter count (6B vs FLUX.2’s larger architecture) for faster inference and lower cost per megapixel (0.005vs0.005 vs 0.012), positioning it for production environments generating thousands of assets. FLUX.2 [dev] offers higher detail fidelity and more sophisticated prompt following for applications where output quality justifies higher per-image costs. FLUX.2 [pro] – Z-Image Turbo delivers 8-step generation at 0.005/MPforrapiditerationworkflows.[FLUX.2[pro]](https://fal.ai/models/fal−ai/flux−2−pro)providesenhancedphotorealismanddetailpreservationat0.005/MP for rapid iteration workflows. [FLUX.2 [pro]](https://fal.ai/models/fal-ai/flux-2-pro) provides enhanced photorealism and detail preservation at 0.03 for the first megapixel, designed for final production assets where maximum quality matters more than generation speed.

​
Capabilities

  • Text prompt input
  • Configurable resolution
  • Adjustable inference steps
  • Reproducible generation (seed)
  • Synchronous mode
  • Batch generation
  • Safety checker

​
API Reference

​
Input Schema

​
prompt
string
required
The prompt to generate an image from.
​
image_size
ImageSize | Enum
default:"landscape_4_3"
The size of the generated image. Default value: landscape_4_3Possible values: square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9
​
num_inference_steps
integer
default:"8"
The number of inference steps to perform. Default value: 8Range: 1 to 8
​
seed
integer
The same seed and the same prompt given to the same version of the model will output the same image every time.
​
sync_mode
boolean
default:"false"
If True, the media will be returned as a data URI and the output data won’t be available in the request history.
​
num_images
integer
default:"1"
The number of images to generate. Default value: 1Range: 1 to 4
​
enable_safety_checker
boolean
default:"true"
If set to true, the safety checker will be enabled. Default value: true
​
output_format
OutputFormatEnum
default:"png"
The format of the generated image. Default value: "png"Possible values: jpeg, png, webp
​
acceleration
AccelerationEnum
default:"regular"
The acceleration level to use. Default value: "regular"Possible values: none, regular, high
​
enable_prompt_expansion
boolean
default:"false"
Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.

​
Output Schema

​
images
list<ImageFile>
required
The generated image files info.
​
timings
Timings
required
The timings of the generation process.
​
seed
integer
required
Seed of the generated Image. It will be the same value of the one passed in the input or the randomly generated that was used in case none was passed.
​
has_nsfw_concepts
list<boolean>
required
Whether the generated images contain NSFW concepts.
​
prompt
string
required
The prompt used for generating the image.

​
Input Example

{
  "prompt": "A hyper-realistic, close-up portrait of a tribal elder from the Omo Valley, painted with intricate white chalk patterns and adorned with a headdress made of dried flowers, seed pods, and rusted bottle caps. The focus is razor-sharp on the texture of the skin, showing every pore, wrinkle, and scar that tells a story of survival. The background is a blurred, smoky hut interior, with the warm glow of a cooking fire reflecting in the subject's dark, soulful eyes. Shot on a Leica M6 with Kodak Portra 400 film grain aesthetic.",
  "image_size": "landscape_4_3",
  "num_inference_steps": 8,
  "sync_mode": false,
  "num_images": 1,
  "enable_safety_checker": true,
  "output_format": "png",
  "acceleration": "regular",
  "enable_prompt_expansion": false
}

​
Output Example

{
  "images": [
    {
      "content_type": "image/png",
      "height": 768,
      "url": "https://storage.googleapis.com/falserverless/example_outputs/z-image-turbo-output.png",
      "width": 1024
    }
  ],
  "prompt": ""
}

​
Limitations

  • image_size restricted to: square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9
  • num_inference_steps range: 1 to 8
  • num_images range: 1 to 4
  • output_format restricted to: jpeg, png, webp
  • acceleration restricted to: none, regular, high
  • Content moderation via safety checker