GLM (Z.ai)
Zhipu's GLM model family, offered as a chat assistant, an agent platform and open weights.
GLM is the model family from Zhipu, reachable as a hosted assistant, an agent and coding platform, and as open weights you can run yourself. The breadth is the point: the same family covers casual assistant use, agent workflows that produce research reports or slide decks, and self-hosted deployment for work that cannot leave your infrastructure. The international product is Z.ai and the mainland one is Zhipu Qingyan; they are separate services with different feature sets, so the right entry point depends on where you are.
Best for
- General assistant and writing work
- Agent and coding tasks on an open base
- Teams that need a self-hostable option
Key features
- Assistant with agent, research-report and slide-generation modes
- Coding-oriented variants for development work
- Open-weight releases for local deployment
Why it matters
It covers the full stack — hosted assistant, agent platform and downloadable weights — from one model family.
