首頁 › Z-Image
Z-Image Prompts
Z-Image is an efficient open-weight text-to-image model from Alibaba's Tongyi lab; the Turbo variant reaches near-flagship quality in very few inference steps, making it fast and cheap to run, with solid bilingual text rendering. Every sample image on this site is generated with Z-Image Turbo — so all prompts here are field-tested: copy one and you reproduce exactly what you see.
Product 奢華香水瓶棚拍
Product 漂浮花瓣面霜罐
Product 球鞋潑濺動態拍攝
Product 奢華腕錶微距棚拍
Product 無線耳機漂浮概念
Product 汽水罐冰塊飛濺
Product 保養精華滴管瓶
Product 鑽戒珠寶微距拍攝
Product 煙霧中漂浮的智慧型手機
Product 高級包裝盒棚拍展示
Product 智慧手錶主打棚拍
Product 智慧手錶生活風格細節微距
Product 球鞋漂浮主打棚拍
Product 高級皮革球鞋棚拍
Product 無線耳機充電盒棚拍
Product 耳罩式耳機漂浮主打
Product 香水瓶水花主打
Product 黑底琥珀香水潑濺
Product 保養精華滴管棚拍
Product 精華瓶漂浮潑濺
Product 設計師太陽眼鏡主打棚拍
Product 反光表面上的太陽眼鏡
Product 無反相機主打棚拍
Product 復古相機漂浮展示
FAQ
Is Z-Image Turbo quality good enough?
Yes — Turbo hits near-flagship perceived quality in few steps. The thousand-plus sample images across this site are all Z-Image Turbo output; judge for yourself on any prompt page.
Anything special about prompting Z-Image?
Same structure as mainstream models: subject + style + lighting + lens + details. English prompts are steadiest, Chinese works too. The steps/cfg/size on each page are our tested values.
Where can I use Z-Image?
It is open-weight — deploy locally in ComfyUI or use cloud APIs. Or just use this site's free generator, which runs Z-Image Turbo.