GLM-4-9B-Chat

GLM-4-9B-Chat is the open-source chat version of Zhipu AI's GLM-4 series, released in June 2024 (THUDM/glm-4-9b-chat). The 9B dense model supports multi-turn dialogue, web browsing, code execution, custom tool calls, and long-text reasoning with up to 128K context, with a 1M-context variant released alongside it. At launch it was a strong small-model option for agent-style workloads on consumer hardware, and Zhipu positioned it as the practical open tier beneath its flagship API models.

Limitations: the original 2024 release is governed by Zhipu's custom non-OSI 'glm-4' license — commercial use is registration-gated, the grant is revocable, and it carries field-of-use restrictions, so it is not equivalent to an Apache/MIT release. Zhipu later moved the GLM line to MIT licensing (GLM-4-32B-0414 onward), which supersedes this checkpoint for new work. It also ships no first-party safety guard and its training data and code are closed, so teams should evaluate alignment behavior themselves.