Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | xAI |
|---|---|
| Context | 500K |
| License | |
| Cheapest hosted (out/1M) | $6.00 · xAI |
| Released | Jul 8, 2026 |
| LiveBench | 75.77 |
| LiveBench Agentic Coding | 56.46 |
| LiveBench Coding | 68.59 |
| LiveBench Data Analysis | 73.04 |
| LiveBench Instruction Following | 71.53 |
| LiveBench Language | 82.8 |
| LiveBench Mathematics | 90.82 |
| LiveBench Reasoning | 87.17 |
| Arena Agent | 22nd of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 13th of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 14th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 27th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 9th of 55 |
| Arena Coding | 1513 |
| Arena Creative Writing | 1452 |
| Arena Hard Prompts | 1489 |
| Arena Instruction Following | 1462 |
| Arena Maths | 1471 |
| Arena Text (overall) | 1465 |
| Arena Code (WebDev) | 1552 |
