Claude Opus 4.6 vs Claude Sonnet 4.6
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | Claude Opus 4.6 | Claude Sonnet 4.6 |
|---|---|---|
| Provider |  Anthropic | Anthropic |
| pontuação média | 57.5 | 47.6 |
| Preço de entrada | $5.00 | $3.00 |
| Preço de saída | $25.00 | $15.00 |
| Janela de contexto | 1.0M tokens (~500 books) | 1.0M tokens (~500 books) |
| Lançado em | 2026-02-04 | 2026-02-17 |
| Código aberto | Proprietary | Proprietary |
Pontuações de benchmark
13 benchmarks · Claude Opus 4.6: 13, Claude Sonnet 4.6: 0
| Benchmark | Categoria | Claude Opus 4.6 | Claude Sonnet 4.6 |
|---|---|---|---|
| ARC-AGI | reasoning | 94.0 | 86.5 |
| ARC-AGI-2 | reasoning | 69.2 | 60.4 |
| Chatbot Arena Elo — Coding | arena | 1542.9 | 1521.0 |
| Chatbot Arena Elo — Overall | arena | 1496.6 | 1462.2 |
| Chess Puzzles | knowledge | 17.0 | 13.0 |
| FrontierMath-2025-02-28-Private | math | 40.7 | 32.4 |
| FrontierMath-Tier-4-2025-07-01-Private | math | 22.9 | 8.3 |
| GPQA diamond | knowledge | 87.4 | 83.2 |
| OTIS Mock AIME 2024-2025 | math | 94.4 | 85.8 |
| PostTrainBench | knowledge | 23.2 | 16.4 |
| SimpleQA Verified | knowledge | 46.5 | 29.0 |
| SWE-Bench verified | coding | 78.7 | 75.2 |
| WeirdML | coding | 77.9 | 66.1 |