Skip to main content
MindStudio
Pricing
BlogAbout
My Workspace
Image Generation Model

Z Image Turbo Controlnet

Z Image Turbo Controlnet is an image generation model from Qwen that uses ControlNet to guide outputs from a reference image.

PublisherQwen
TypeImage
Context Window10,000 tokens
ReleasedNovember 2025
Price$0.012/image
ProviderWaveSpeed
Source Image

ControlNet-guided image generation with source images

Z Image Turbo Controlnet is an image generation model published by Qwen and served through WaveSpeed AI. It uses ControlNet conditioning, which allows a reference image to guide the structure, pose, or composition of generated outputs rather than relying on text prompts alone. The model accepts an image URL as input alongside a selectable ControlNet mode, making it suited for workflows where spatial or structural consistency with a source image is required.

The model is priced at $0.012 per image and was released in November 2025. It supports configurable output size, a numeric strength parameter that controls how closely the output adheres to the reference image, and a seed input for reproducible results. These controls make it practical for tasks like style transfer, pose-consistent character generation, and layout-constrained image creation.

What Z Image Turbo Controlnet supports

ControlNet Conditioning

Uses a reference image to guide the structure or composition of generated outputs via a selectable ControlNet mode. This allows spatial constraints like pose or edges to be preserved in the result.

Reference Image Input

Accepts an image URL as a direct input to anchor generation to a source image. This enables layout- or structure-consistent image creation without relying solely on text prompts.

Strength Control

A numeric strength parameter determines how closely the generated image adheres to the reference. Lower values allow more creative deviation; higher values enforce tighter structural fidelity.

Configurable Output Size

Output image dimensions are selectable via a size input, allowing the model to fit different aspect ratios or resolution requirements for a given project.

Reproducible Generation

A seed input lets users lock in a specific random state so that identical prompts and settings produce the same image output across runs.

Ready to build with Z Image Turbo Controlnet?

Get Started Free

Common questions about Z Image Turbo Controlnet

How much does Z Image Turbo Controlnet cost to use?

The model is priced at $0.012 per image generated.

What is the context window for this model?

The model has a context window of 10,000 tokens.

What inputs does Z Image Turbo Controlnet accept?

It accepts a reference image URL, a ControlNet mode selection, an output size selection, a numeric strength value, and a seed for reproducibility.

What does the strength parameter do?

The strength parameter controls how closely the generated image follows the reference image. A higher value enforces tighter adherence to the source image's structure, while a lower value allows more variation.

When was Z Image Turbo Controlnet released?

The model was released in November 2025 and became available on MindStudio on March 25, 2026.

Parameters & options

Reference ImageImage URL

Reference image URL for ControlNet to extract structural guidance from.

ControlNet ModeSelect

ControlNet mode: 'depth' for depth map guidance, 'canny' for edge detection, 'pose' for human pose estimation, 'none' for no control.

Default: depth
DepthCannyPoseNone
SizeSelect

Output image size in pixels (width*height).

Default: 1024*1024
1024×1024 (Square)1024×1536 (Portrait)1536×1024 (Landscape)768×10241024×768768×13441344×768512×512
StrengthNumber

Controls how strongly the ControlNet guidance affects the output. Higher values follow the control signal more strictly.

Default: 0.7Range: 0–1 (step 0.05)
SeedSeed

Random seed for reproducible generation. Use -1 for random seed.

Range: -1–2147483647

Start building with Z Image Turbo Controlnet

No API keys required. Create AI-powered workflows with Z Image Turbo Controlnet in minutes — free.