Z-Image Turbo model guide
A practical guide to Z-Image Turbo on GPT Image Hub: Alibaba's fast, low-cost text-to-image model at 1K and 2K, with measured render times and real photographic examples.
Z-Image Turbo is Alibaba Tongyi's fast text-to-image model. On GPT Image Hub it renders 1K and 2K images for 15 credits each, at Alibaba's full list price, in any of the Hub's 16 aspect ratios. In the Hub's tests a 2K product photo took 11 seconds and 1K scenes took 42 to 70 seconds. It does not accept reference images, so use another model for edits.
Visual examples
Visual examples
A neon ramen stall in a rainy Tokyo alley.
Night street scene
A neon ramen stall in a rainy Tokyo alley.
Neon-lit ramen stall in a rainy Tokyo alley at night.Cafe close-up
Latte art in morning light at 1K.
Ceramic mug with latte art on a marble table.2K product in 11 seconds
A clean product photo at 2K for the same 15 credits.
Matte black water bottle on a light oak table.Best use cases
- Photographic scenes on a budget.
- Product photos at 2K.
- Food and cafe imagery.
- Moody street and night scenes.
Where to be careful
- Editing an existing image; Z-Image Turbo is text-to-image only.
- 4K output; it renders up to 2K on the Hub.
Strengths
Low price: 15 credits at 1K or 2K.
Photorealistic food, product, and street scenes.
All 16 aspect ratios on the Hub, from 1:3 to 3:1.
Fast 2K: 11 seconds for a product photo in Hub tests.
Use cases
Budget photography
Realistic scenes for blogs and social posts at 15 credits.
Product shots
Clean 2K product photos in seconds.
Food and cafe
Appetizing close-ups with natural light.
Workflow
Write a photo caption
Describe subject, light, and lens.
Render at 2K
Same price as 1K, often faster.
Edit elsewhere
Move to a reference-capable model for edits.
Prompting guidance
- Write prompts like photo captions: subject, light, lens, mood.
- Add "no text" for clean images.
- Use 2K for product shots; it costs the same as 1K.
- Describe weather and reflections for night scenes.
Copy-ready prompts
Cinematic wide shot of a neon-lit ramen stall in a rainy Tokyo alley at night, reflections on wet pavement, no text.
Close-up photo of a ceramic coffee mug with latte art on a marble cafe table, morning sunlight, steam rising, shallow depth of field, no text.
Minimal product photo of a matte black water bottle on a light oak table, soft daylight, no text.
Example outputs generated on GPT Image Hub
Official sources
FAQ
Can Z-Image Turbo edit images?
No. Z-Image Turbo is text-to-image only, so the Hub does not accept reference images for it. Use Seedream or Nano Banana models for edits.
How much does Z-Image Turbo cost?
15 credits per image (about $0.015) at 1K or 2K, based on Alibaba Cloud Model Studio's ¥0.1 list price.
How fast is Z-Image Turbo?
In the Hub's tests a 2K product photo took 11 seconds and 1K scenes took 42 to 70 seconds.
Use this model
Start from a prompt template or compare models first
Use the generator when you already know the model, or compare all supported models when the prompt needs a different balance of speed, detail, and reference inputs.