Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | Meta |
|---|---|
| Context | 1M |
| License | |
| Cheapest hosted (out/1M) | $4.25 · Meta |
| Released | Aug 5, 2026 |
| LiveBench | 77.95 |
| LiveBench Agentic Coding | 57.58 |
| LiveBench Coding | 77.54 |
| LiveBench Data Analysis | 76.46 |
| LiveBench Instruction Following | 74.33 |
| LiveBench Language | 78.57 |
| LiveBench Mathematics | 91.2 |
| LiveBench Reasoning | 90 |
| Arena Agent | 37th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 12th of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 43rd of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 37th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 28th of 55 |
| Arena Coding | 1535 |
| Arena Creative Writing | 1448 |
| Arena Hard Prompts | 1511 |
| Arena Instruction Following | 1475 |
| Arena Text (overall) | 1496 |
| Arena Code (WebDev) | 1531 |
