Claude Opus 4.6 vs Gemini 3.1 Pro Preview
Côte à côte. Chaque métrique. Chaque benchmark.
| Type | Claude Opus 4.6 | Gemini 3.1 Pro Preview |
|---|---|---|
| Provider |  Anthropic |  Google DeepMind |
| score moyen | 57.5 | 60.6 |
| Prix d'entrée | $5.00 | $2.00 |
| Prix de sortie | $25.00 | $12.00 |
| Fenêtre de contexte | 1.0M tokens (~500 books) | 1.0M tokens (~524 books) |
| Sorti le | 2026-02-04 | 2026-02-19 |
| Code source ouvert | Proprietary | Proprietary |
Scores de benchmark
16 benchmarks · Claude Opus 4.6: 7, Gemini 3.1 Pro Preview: 9
| Benchmark | Catégorie | Claude Opus 4.6 | Gemini 3.1 Pro Preview |
|---|---|---|---|
| APEX-Agents | agentic | 31.7 | 33.5 |
| ARC-AGI | reasoning | 94.0 | 98.0 |
| ARC-AGI-2 | reasoning | 69.2 | 77.1 |
| Chatbot Arena Elo — Coding | arena | 1542.9 | 1455.7 |
| Chatbot Arena Elo — Overall | arena | 1496.6 | 1492.6 |
| Chess Puzzles | knowledge | 17.0 | 55.0 |
| FrontierMath-2025-02-28-Private | math | 40.7 | 36.9 |
| FrontierMath-Tier-4-2025-07-01-Private | math | 22.9 | 16.7 |
| GPQA diamond | knowledge | 87.4 | 92.1 |
| OTIS Mock AIME 2024-2025 | math | 94.4 | 95.6 |
| PostTrainBench | knowledge | 23.2 | 21.6 |
| SimpleBench | reasoning | 61.1 | 75.5 |
| SimpleQA Verified | knowledge | 46.5 | 77.3 |
| SWE-Bench verified | coding | 78.7 | 75.6 |
| Terminal Bench | coding | 74.7 | 78.4 |
| WeirdML | coding | 77.9 | 72.1 |