Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | Qwen |
|---|---|
| Context | 1M |
| License | |
| Cheapest hosted (out/1M) | $3.75 · Novita AI |
| Released | May 21, 2026 |
| LiveBench | 73.14 |
| LiveBench Agentic Coding | 43.59 |
| LiveBench Coding | 74.22 |
| LiveBench Data Analysis | 71.79 |
| LiveBench Instruction Following | 74.04 |
| LiveBench Language | 79.74 |
| LiveBench Mathematics | 85.25 |
| LiveBench Reasoning | 83.34 |
| Arena Agent | 40th of 55 |
| Arena Agent · Recoverygets back on track after a command fails | 34th of 55 |
| Arena Agent · Steerabilitydoes what it was asked, and changes course when told | 34th of 55 |
| Arena Agent · Task outcomefinishes what the session set out to do | 43rd of 55 |
| Arena Agent · Tool usereaches for the right one, and does not invent one | 41st of 55 |
| Arena Coding | 1525 |
| Arena Creative Writing | 1441 |
| Arena Hard Prompts | 1496 |
| Arena Instruction Following | 1466 |
| Arena Maths | 1485 |
| Arena Text (overall) | 1475 |
| Arena Code (WebDev) | 1515 |
