GLM-5.3 Flash

GLM-5.3-Flash is Z.AI's first natively multimodal model in the GLM-5 series. It has 320B total parameters, with 18B active, and a hybrid attention design intended to lower the cost of long prompts.

The model can work with images and video, use browser and computer tools, and inspect rendered interfaces. Its 1,048,576-token context window suits coding agents, visual development, and other tasks that need a lot of context. It is cheaper to run than GLM-5.2, while benchmark results show stronger performance across the tests Z.AI reports.