Skip to main content
z-image is an asynchronous text-to-image model. It creates image generation tasks from text prompts and returns a public GENGEN task ID immediately after the task is accepted. The model supports custom width and height from 256 to 1536 pixels and optional prompt rewriting for expanded prompt detail before generation. In GENGEN, z-image uses its own image task endpoint instead of the synchronous Seedream image generation endpoint.

​
Model ID

Pass this exact public ID in the top-level model field. Do not substitute an upstream provider alias.

​
Supported modes

  • text_to_image

​
Inputs

  • Text

​
Outputs

  • Image

​
Capabilities

  • Text-to-image

​
Additional capabilities

  • Text-to-image
  • Custom dimensions
  • Prompt rewriting

​
Technical details

Availability and final customer prices can change. Check the Dashboard model catalog before production use.
See the API reference overview for authentication, request conventions, and response schemas.