CogView4, An Open-Source Text-to-Image Model, Has Been Released by Zhipu AI
On March 4th, the large model unicorn company Zhipu AI released the first model of the 2025 open source year: the first open-source text-to-image model that supports generating Chinese characters – CogView4.
It is reported that the model ranks first in the comprehensive score of DPG-Bench benchmark tests, achieving SOTA (state-of-the-art) among open-source image generation models. This model is also the first image generation model to follow the Apache 2.0 license.
The GitHub page shows that Zhipu AI has adapted and open-sourced the diffusers version of the CogView-4 model, which has 6 billion parameters and supports native Chinese input as well as generating images from Chinese text.
SEE ALSO: Zhipu AI Secures Strategic Financing from Hangzhou State-Owned Investors
