Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | OpenAI |
|---|---|
| Context | 1.1M |
| License | |
| Cheapest hosted (out/1M) | $10.00 · Microsoft Azure AI |
| Released | Sep 22, 2026 |
| LiveBench | 79.25 |
| LiveBench Agentic Coding | 52.88 |
| LiveBench Coding | 81.77 |
| LiveBench Data Analysis | 81.19 |
| LiveBench Instruction Following | 68.57 |
| LiveBench Language | 85.3 |
| LiveBench Mathematics | 96.36 |
| LiveBench Reasoning | 88.65 |
| Arena Agent | 5th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 6th of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 1st of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 4th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 9th of 55 |
| Arena Coding | 1524 |
| Arena Creative Writing | 1451 |
| Arena Hard Prompts | 1485 |
| Arena Instruction Following | 1447 |
| Arena Maths | 1450 |
| Arena Text (overall) | 1457 |
| Arena Code (WebDev) | 1681 |
