GLM 5.2
Chat model · Z.ai ·
GLM 5.2 is a chat model by Z.ai. In Endless it costs 64.8 credits per 1M input tokens and 206.6 per 1M output tokens, on the Hobby plan and up.
GLM 5.2 is Z.ai's coding and tool-use model built for long tasks, with a 1M-token context that fits a whole project and optional thinking.
- Price
- 64.8 credits per 1M input tokens
- 206.6 credits per 1M output tokens
- Availability
- Available on Hobby and above
- Context window
- 1M tokens
- Max output
- 128K tokens
- Inputs
- Text
- Outputs
- Text
About GLM 5.2
GLM 5.2 is a Z.ai model made for long-horizon work. Its 1M-token context is built to hold project-scale engineering material, so a repository's key files, specs and logs can sit in one conversation. It handles function calling and structured JSON output, and you can run it with thinking enabled for careful reasoning or disabled for faster replies.
Developers use it for tasks that sprawl: adding a feature that touches many files, writing tests for an existing module, or chaining tool calls to gather data and act on it. Because the context is so large, it keeps track of decisions made early in a long session instead of drifting.
GLM 5.3 shares the same base model with extra post-training for software engineering and agents, and it always reasons. GLM 5.2 suits you when you want the option of quick answers without thinking. Both live in Endless chat, so switching between them in your workspace takes one click.
Best for
- Feature work that spans many files in a repository
- Writing tests and docs for existing code
- Chained tool calls that collect data and act on it
- Quick coding answers with thinking turned off
Questions about GLM 5.2
- How much code can GLM 5.2 read at once?
- GLM 5.2 holds up to 1M tokens of context, enough for a large share of a codebase plus specs and logs, and it can reply with up to 128K tokens.
- What is the difference between GLM 5.2 and GLM 5.3?
- Both use the same base model. GLM 5.3 adds more post-training for software engineering and agent tasks and always reasons, while GLM 5.2 lets you switch thinking off for faster replies.
Related models
GLM 5.3
by Z.ai
Always-on reasoning for hard software and agent tasks
113.4 / 356.4 credits per 1M tokens
GPT-6.1 Sol
by OpenAI
Near-flagship agentic coding and professional work at lower cost
162 / 810 credits per 1M tokens
Claude Opus 5.5
by Anthropic
Long agentic coding and research work over a 1M-token context
324 / 1620 credits per 1M tokens
Give your next idea a place to grow.
Start free, create a project, and bring your brand’s references together. Or book a demo to explore Endless with your team.