Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | Anthropic |
|---|---|
| Context | 1M |
| Weights | |
| License | — |
| — | |
| Cheapest hosted (out/1M) | $25.00 · Microsoft Azure AI |
| Released | Feb 4, 2026 |
| LiveBench | 74.5via Thinking Auto High effort |
| LiveBench Agentic Coding | 49via Thinking Auto High effort |
| LiveBench Coding | 78.2via Thinking Auto High effort |
| LiveBench Data Analysis | 69.9via Thinking Auto High effort |
| LiveBench Instruction Following | 63.3via Thinking Auto High effort |
| LiveBench Language | 83.3via Thinking Auto High effort |
| LiveBench Mathematics | 89.3via Thinking Auto High effort |
| LiveBench Reasoning | 88.7via Thinking Auto High effort |
| Arena Agent (IPS) | 0.065 |
| Arena Coding | 1547.9 |
| Arena Creative Writing | 1478 |
| Arena Hard Prompts | 1526.7 |
| Arena Instruction Following | 1499.4 |
| Arena Maths | 1504.5 |
| Arena Text (overall) | 1497.2 |
| Arena Code (WebDev) | 1537.9 |
| SWE-bench Verified | 75.6via mini-SWE-agent |
