Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | Anthropic |
|---|---|
| Context | 1M |
| License | |
| Cheapest hosted (out/1M) | $50.00 · Anthropic |
| Released | Sep 1, 2026 |
| LiveBench | 83.41 |
| LiveBench Agentic Coding | 66.06 |
| LiveBench Coding | 86.38 |
| LiveBench Data Analysis | 80.28 |
| LiveBench Instruction Following | 72.99 |
| LiveBench Language | 89.5 |
| LiveBench Mathematics | 97.01 |
| LiveBench Reasoning | 91.69 |
| Arena Agent | 1st of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 2nd of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 6th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 1st of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 24th of 55 |
| Arena Coding | 1529 |
| Arena Creative Writing | 1484 |
| Arena Hard Prompts | 1520 |
| Arena Instruction Following | 1497 |
| Arena Maths | 1517 |
| Arena Text (overall) | 1501 |
| Arena Code (WebDev) | 1751 |
