MODELS · IMAGE · ALIBABA
Z-Image.
Apache 2.0 open-source model from Alibaba — the full 6B Z-Image, built for creative control, not just speed
Free to start · 100+ models on one account · cancel anytime
About Z-Image
Z-Image is Alibaba's open-source, Apache 2.0-licensed image generation model — a 6-billion-parameter Scalable Single-Stream Diffusion Transformer for text-to-image and image-to-image generation. This is the full, undistilled version: Alibaba also publishes a faster, distilled Z-Image-Turbo that trims generation down to 8 steps, but drops classifier-free guidance, negative prompting, and some output diversity in exchange for that speed. The full model keeps all three — negative prompts, adjustable guidance, and 28 to 50 generation steps — for finer control over the final image, not just a fast draft.
- Text to image
- Image to image
How to use Z-Image on getvivix
Create a free getvivix account — no card required.
Choose Z-Image from the model list and set your options.
Enter your prompt or upload your input, hit generate, then download in full quality.
Z-Image — frequently asked
Z-Image is one of 100+ AI models available on getvivix. Z-Image is Alibaba's open-source, Apache 2.0-licensed image generation model — a 6-billion-parameter Scalable Single-Stream Diffusion Transformer for text-to-image and image-to-image generation. This is the full, undistilled version: Alibaba also publishes a faster, distilled Z-Image-Turbo that trims generation down to 8 steps, but drops classifier-free guidance, negative prompting, and some output diversity in exchange for that speed. The full model keeps all three — negative prompts, adjustable guidance, and 28 to 50 generation steps — for finer control over the final image, not just a fast draft.
Sign in to getvivix and open the Studio, pick Z-Image from the model list, enter your prompt (or upload your input), and generate — then download the result in full quality.
Yes — getvivix has a free tier, so you can try Z-Image without a card. Sign up and start generating right away, alongside 100+ other AI models on one account.
Z-Image supports text to image, image to image. It runs on getvivix alongside 100+ other frontier AI models, all from one account.