Z-Image Turbo
Z-Image Turbo supports text-to-image generation and image-to-image transformation. It produces photorealistic imagery from written descriptions and can render text within an image. Its text generation capabilities include understanding English and Chinese prompts.
An existing image can serve as the starting point for a transformation guided by a written description. The degree of change is adjustable: lighter transformations support detail refinement, while stronger transformations allow broader changes to style or content. This gives designers a way to move from an initial concept into visual revisions using the same model.
What it's good for
Text-led generation suits product-photo concepts, campaign banners, and poster layouts with embedded lettering. Image-to-image work suits exploring artistic treatments or revising a composition that already has a useful structure. Together, these tasks support a workflow that begins with a written brief and continues through variations of the selected image.
