Alibaba's Tongyi-MAI team has released Z-Image-Turbo, an open-source text-to-image model that generates photorealistic outputs in just 8 diffusion steps with sub-second latency. Ranked #1 among open-source models globally, it combines breakthroughs like Decoupled-DMD distillation and bilingual text rendering to make professional-grade image synthesis accessible on consumer GPUs.