无限量使用 Z-Image Turbo，无每日限制。阿里巴巴通义实验室的60亿参数开源模型采用 S3-DiT 架构和 Decoupled-DMD 蒸馏技术，仅需8步推理即可实现亚秒级生成。在开源图像生成器中排名第一，具有出色的文字渲染能力。
观看社区教程和评测，了解 Z-Image Turbo 的快速图像生成能力
看看大家在 X (Twitter) 上对 Z-Image Turbo 的评价
Z-Image Turbo is the new #1 open weights Text to Image model, surpassing FLUX.2 [dev], HunyuanImage 3.0 (Fal), and Qwen-Image in the Artificial Analysis Image Arena. Z-Image Turbo is the first model released by the Alibaba Tongyi-MAI team (separate from the Alibaba Wan, and Qwen Show more
A huge thank you to the community for the recognition. 🧡 Seeing Z-Image Turbo ranked as the top open-weights model by @ArtificialAnlys is a proud moment for the Alibaba Tongyi-MAI team. We designed this 6B model to be both powerful and affordable, and we are so glad it is Show more
Z-Image Turbo is the new #1 open weights Text to Image model, surpassing FLUX.2 [dev], HunyuanImage 3.0 (Fal), and Qwen-Image in the Artificial Analysis Image Arena. Z-Image Turbo is the first model released by the Alibaba Tongyi-MAI team (separate from the Alibaba Wan, and Qwen
Z Image Turbo is now available! This model from Alibaba\'s Tongyi-MAI delivers photorealistic images in <1s for <$0.01. It handles both English and Chinese and can render multilingual text directly in images - great for posters, product shots, and cross-market campaigns.
Introducing⚡️Z-Image ("造相") from Alibaba Tongyi Lab! Our highly-efficient 6B-parameter image generation model, built on a novel Single-Stream DiT architecture, is the best & fastest open image model with stunning quality. Explore more in our blog: tongyi-mai.github.io/Z-Image-blog/ 1/n
Stop juggling five different ControlNet files. Alibaba-PAI just dropped a Z-Image-Turbo ControlNet Union that handles Canny, Depth, Pose, HED, and MLSD in a single model. Trained on 1 million images at 1328px resolution, so it actually respects high-res details.
Z-Image: Alibaba’s efficient 6B model is nearly here. Photorealistic output, superior bilingual text rendering; precise instruction-based editing. ComfyUI 0-day support modelscope.cn/models/Tongyi-…
Z-Image Turbo Controlnet is here! A multi-condition ControlNet - Canny, HED, Depth, Pose, and MLSD; huggingface.co/alibaba-pai/Z-…
Z-IMAGE TURBO is a *serious* competitor to FLUX.2 A hyper-realistic, close-up portrait of a age 40 rugged warrior mixed heritage norwegian bearded axe warrior wearing a ragged cloak and cobbled armor, standing by a dragon . Natural Light. Shot on a Leica M6 with Kodak Portra Show more
体验 Z-Image Turbo 的超快图像生成。在3秒内生成令人惊叹的图像，涵盖多种风格——从电影美学到详细的产品渲染和艺术插画，品质卓越。
捕捉王家卫电影的氛围感，包含情绪化的灯光、城市场景和怀旧的1990年代香港美学。
“Wong Kar-wai film style, a lonely man smoking a cigarette in a narrow Hong Kong hallway, 1990s...”
阿里巴巴通义实验室的60亿参数开源文生图模型，实现亚秒级生成
Z-Image Turbo 采用 S3-DiT 架构和 Decoupled-DMD 蒸馏技术，仅需8步推理即可生成逼真图像。在开源图像生成器中排名第一，具有出色的中英文文字渲染能力。
阿里巴巴通义实验室的60亿参数开源文生图模型，实现亚秒级生成
Z-Image Turbo 采用 S3-DiT 架构和 Decoupled-DMD 蒸馏技术，仅需8步推理即可生成逼真图像。在开源图像生成器中排名第一，具有出色的中英文文字渲染能力。
使用阿里巴巴超快速AI模型即时生成图像
