Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | Z.ai |
|---|---|
| Context | 1M |
| License | |
| Cheapest hosted (out/1M) | $2.28 · InferenceNet |
| Released | Aug 18, 2026 |
| LiveBench | 76.14 |
| LiveBench Agentic Coding | 60.91 |
| LiveBench Coding | 78.95 |
| LiveBench Data Analysis | 70.24 |
| LiveBench Instruction Following | 69.3 |
| LiveBench Language | 79.86 |
| LiveBench Mathematics | 87.9 |
| LiveBench Reasoning | 85.8 |
| Arena Agent | 17th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 32nd of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 17th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 13th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 9th of 55 |
| Arena Coding | 1522 |
| Arena Creative Writing | 1456 |
| Arena Hard Prompts | 1506 |
| Arena Instruction Following | 1480 |
| Arena Maths | 1497 |
| Arena Text (overall) | 1480 |
| Arena Code (WebDev) | 1619 |
