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) | $9.00 · Google AI |
| Released | May 19, 2026 |
| LiveBench | 74.64 |
| LiveBench Agentic Coding | 48.99 |
| LiveBench Coding | 78.18 |
| LiveBench Data Analysis | 64.86 |
| LiveBench Instruction Following | 75.6 |
| LiveBench Language | 84.58 |
| LiveBench Mathematics | 88.24 |
| LiveBench Reasoning | 82 |
| Arena Agent | 38th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 43rd of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 35th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 29th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 33rd of 55 |
| Arena Coding | 1507 |
| Arena Creative Writing | 1466 |
| Arena Hard Prompts | 1495 |
| Arena Instruction Following | 1465 |
| Arena Maths | 1496 |
| Arena Text (overall) | 1477 |
| Arena Code (WebDev) | 1499 |
