Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | Google |
|---|---|
| Context | 1M |
| License | |
| Cheapest hosted (out/1M) | $3.75 · Google AI |
| Released | Sep 2, 2026 |
| LiveBench | 75.83 |
| LiveBench Agentic Coding | 54.24 |
| LiveBench Coding | 72.49 |
| LiveBench Data Analysis | 54.01 |
| LiveBench Instruction Following | 81.41 |
| LiveBench Language | 87.79 |
| LiveBench Mathematics | 91.56 |
| LiveBench Reasoning | 89.29 |
| Arena Agent | 12th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 31st of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 15th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 6th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 31st of 55 |
| Arena Coding | 1532 |
| Arena Creative Writing | 1482 |
| Arena Hard Prompts | 1516 |
| Arena Instruction Following | 1488 |
| Arena Maths | 1522 |
| Arena Text (overall) | 1492 |
| Arena Code (WebDev) | 1580 |
