Z-Image Turbo
AI imagesZ-Image Turbo is an ultra-fast 6B-parameter text-to-image model developed by Tongyi-MAI that can generate high-quality images in seconds. It features photorealistic capabilities, can accurately render bilingual Chinese and English text, has a built-in prompt enhancement engine, and can achieve sub-second inference on an H800 GPU, about 2-3 seconds on consumer-grade RTX 3090/4090, producing high-quality images in just 8 steps. The Z-Image series includes three variants: the Turbo distilled version, the Base foundation version, and the Edit image editing version, and supports commercial use (Apache-2.0 license) and self-hosting. This tool is suitable for scenarios such as e-commerce product images, bilingual marketing posters, social media visual content, portrait avatars, and food and lifestyle. It currently does not support editing or modifying uploaded images, prohibits the generation of NSFW content, requires a GPU with at least 16GB of VRAM for local operation, and allows a maximum of 4 images per API batch request.