cogview-4-t2i

CogView-4 is an open-source text-to-image model developed by Tsinghua University/Zhipu AI, based on diffusion technology. It supports both Chinese and English, can generate high-resolution images, and demonstrates leading semantic alignment performance. The latest version 0304 has been released as open source, providing a diffusers implementation with approximately 6 billion parameters, natively supporting Chinese text-to-image generation and Chinese character rendering.