Gemini 3.1 Pro Preview vs Claude Opus 4.6
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | Gemini 3.1 Pro Preview | Claude Opus 4.6 |
|---|---|---|
| Provider |  Google DeepMind |  Anthropic |
| pontuação média | 60.6 | 57.5 |
| Preço de entrada | $2.00 | $5.00 |
| Preço de saída | $12.00 | $25.00 |
| Janela de contexto | 1.0M tokens (~524 books) | 1.0M tokens (~500 books) |
| Lançado em | 2026-02-19 | 2026-02-04 |
| Código aberto | Proprietary | Proprietary |
Pontuações de benchmark
16 benchmarks · Gemini 3.1 Pro Preview: 9, Claude Opus 4.6: 7
| Benchmark | Categoria | Gemini 3.1 Pro Preview | Claude Opus 4.6 |
|---|---|---|---|
| APEX-Agents | agentic | 33.5 | 31.7 |
| ARC-AGI | reasoning | 98.0 | 94.0 |
| ARC-AGI-2 | reasoning | 77.1 | 69.2 |
| Chatbot Arena Elo — Coding | arena | 1455.7 | 1542.9 |
| Chatbot Arena Elo — Overall | arena | 1492.6 | 1496.6 |
| Chess Puzzles | knowledge | 55.0 | 17.0 |
| FrontierMath-2025-02-28-Private | math | 36.9 | 40.7 |
| FrontierMath-Tier-4-2025-07-01-Private | math | 16.7 | 22.9 |
| GPQA diamond | knowledge | 92.1 | 87.4 |
| OTIS Mock AIME 2024-2025 | math | 95.6 | 94.4 |
| PostTrainBench | knowledge | 21.6 | 23.2 |
| SimpleBench | reasoning | 75.5 | 61.1 |
| SimpleQA Verified | knowledge | 77.3 | 46.5 |
| SWE-Bench verified | coding | 75.6 | 78.7 |
| Terminal Bench | coding | 78.4 | 74.7 |
| WeirdML | coding | 72.1 | 77.9 |