Claude Sonnet 4.6 vs Claude Opus 4.5
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | Claude Sonnet 4.6 | Claude Opus 4.5 |
|---|---|---|
| Provider |  Anthropic | Anthropic |
| pontuação média | 47.6 | 45.4 |
| Preço de entrada | $3.00 | $5.00 |
| Preço de saída | $15.00 | $25.00 |
| Janela de contexto | 1.0M tokens (~500 books) | 200K tokens (~100 books) |
| Lançado em | 2026-02-17 | 2025-11-24 |
| Código aberto | Proprietary | Proprietary |
Pontuações de benchmark
13 benchmarks · Claude Sonnet 4.6: 8, Claude Opus 4.5: 5
| Benchmark | Categoria | Claude Sonnet 4.6 | Claude Opus 4.5 |
|---|---|---|---|
| ARC-AGI | reasoning | 86.5 | 80.0 |
| ARC-AGI-2 | reasoning | 60.4 | 37.6 |
| Chatbot Arena Elo — Coding | arena | 1521.0 | 1465.2 |
| Chatbot Arena Elo — Overall | arena | 1462.2 | 1467.7 |
| Chess Puzzles | knowledge | 13.0 | 12.0 |
| FrontierMath-2025-02-28-Private | math | 32.4 | 20.7 |
| FrontierMath-Tier-4-2025-07-01-Private | math | 8.3 | 4.2 |
| GPQA diamond | knowledge | 83.2 | 81.4 |
| OTIS Mock AIME 2024-2025 | math | 85.8 | 86.1 |
| PostTrainBench | knowledge | 16.4 | 17.3 |
| SimpleQA Verified | knowledge | 29.0 | 41.8 |
| SWE-Bench verified | coding | 75.2 | 76.7 |
| WeirdML | coding | 66.1 | 63.7 |