We’re excited to add GLM-5.2 from Z-AI to the Chat-O model library. It’s a large-scale reasoning model built for long-horizon agent workflows and shipped with a massive 1,000,000-token context window — enough for entire codebases, long documentation sets, and extended multi-step agent tasks.

What’s new in GLM-5.2

• 1M-token context — roughly 5× more than GLM-5.1’s 200K window
• Long-horizon agent workflows — designed to stay coherent across many tool calls and steps
• Strong coding performance — built on the GLM-5 lineage that already topped open coding benchmarks
• Text in / text out — focused, fast, and predictable for agentic use

Available tiers

Both tiers point at the same z-ai/glm-5.2 backend — the tier only changes how credits are billed, not the model quality.

Why we added it

GLM-5.2 fills a gap in our lineup: a reasoning-strong model with a context window long enough to keep an entire project in view at once. That makes it a natural fit for tasks like:

• Loading a whole repository and asking architectural questions
• Debugging across many files in a single conversation
• Long multi-step agent runs where shorter-context models lose the thread

Try it now

Pick GLM-5.2 from the model dropdown in any chat. New users get 1,000 free credits to explore it. No API keys, no setup — just open a chat and start asking.

Already using GLM-5? GLM-5.2 keeps the same familiar interface but with a much larger context window and stronger long-horizon behavior. Both models remain available so you can pick the right one per task.