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 | Jul 21, 2026 |
| LiveBench | 73.59 |
| LiveBench Agentic Coding | 43.43 |
| LiveBench Coding | 77.86 |
| LiveBench Data Analysis | 63 |
| LiveBench Instruction Following | 75.37 |
| LiveBench Language | 83.9 |
| LiveBench Mathematics | 86.4 |
| LiveBench Reasoning | 85.15 |
| Arena Agent | 36th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 37th of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 40th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 28th of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 1st of 55 |
| Arena Coding | 1523 |
| Arena Creative Writing | 1473 |
| Arena Hard Prompts | 1503 |
| Arena Instruction Following | 1477 |
| Arena Maths | 1518 |
| Arena Text (overall) | 1484 |
| Arena Code (WebDev) | 1536 |
