GLM 5.3 now available on AI Gateway
By Steven Van ·
GLM 5.3 from Z.ai is now available on AI Gateway. GLM 5.3 has improvements vs.
Z.ai's GLM 5.3 is now available on Vercel's AI Gateway. Z.ai reports gains over GLM 5.2 on complex software engineering and multi-step agent tasks, reached with fewer output tokens at the same effort level, plus stronger vulnerability discovery that reasons across the stages of an exploitation chain. DeepsecBench, which scores how reliably a model finds security vulnerabilities in application code and what a run costs, now covers GLM 5.3 at both high and max effort.
The model takes text input with a 1M token context window and a 128K token maximum output, both unchanged from GLM 5.2, and supports function calling, structured output, streaming, and context caching. It's available with the slug zai/glm-5.3, and can be connected to coding agents such as Claude Code, Codex, OpenCode, Cursor, and Pi by running vercel ai-gateway coding-agents setup and selecting it from within the agent.
Read the full announcement.
