Z.ai · Model release
GLM 5.3: Coding Plan default and separately billed API
GLM 5.3 became the Coding Plan default in August 2026. The hosted token API and open-weight checkpoint have separate access and licensing terms; check Z.ai's current docs for live rates and limits.
What changed
- Z.ai's Coding Plan announcement states that 5.2/5.1 requests on that plan route to GLM 5.3; this is a plan routing rule, not a statement that API IDs resolve the same way.
- Official docs: 1M context, 128K max output, thinking always on, reasoning_effort low / high / max (default max).
- The separately billed token API has its own live rate table; verify current prices and cache rates at checkout.
Best fit
- Existing GLM Coding Plan teams
- Hosted glm-5.3 token API
- Long-horizon coding in supported harnesses
What to be careful about
- Coding Plan usage and token-API billing are separate; compare the applicable contract and current rate table.
- Vendor benchmark claims use vendor-selected evaluation setups; do not compare them directly with a public board without matching harness and settings.
- Thinking cannot be disabled; a `disabled` request fails.
Dataset snapshot: GLM 5.3 is also represented in the checked comparison dataset. Prices and benchmark figures carry their own field dates.