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) | $15.00 · OpenAI |
| Released | Mar 5, 2026 |
| LiveBench | 77.97 |
| LiveBench Agentic Coding | 53.84 |
| LiveBench Coding | 77.54 |
| LiveBench Data Analysis | 79.31 |
| LiveBench Instruction Following | 70.22 |
| LiveBench Language | 82.63 |
| LiveBench Mathematics | 94.15 |
| LiveBench Reasoning | 88.12 |
| Arena Agent | 26th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 21st of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 20th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 34th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 9th of 55 |
| Arena Coding | 1511 |
| Arena Creative Writing | 1434 |
| Arena Hard Prompts | 1487 |
| Arena Instruction Following | 1460 |
| Arena Maths | 1460 |
| Arena Text (overall) | 1465 |
| Arena Code (WebDev) | 1394 |
