gpt-oss-120b vs Claude Opus 4.5
Côte à côte. Chaque métrique. Chaque benchmark.
| Type | gpt-oss-120b | Claude Opus 4.5 |
|---|---|---|
| Provider |  OpenAI |  Anthropic |
| score moyen | 46.9 | 45.4 |
| Prix d'entrée | $0.04 | $5.00 |
| Prix de sortie | $0.19 | $25.00 |
| Fenêtre de contexte | 131K tokens (~66 books) | 200K tokens (~100 books) |
| Sorti le | 2025-08-05 | 2025-11-24 |
| Code source ouvert | Open Source | Proprietary |
Scores de benchmark
11 benchmarks · gpt-oss-120b: 2, Claude Opus 4.5: 9
| Benchmark | Catégorie | gpt-oss-120b | Claude Opus 4.5 |
|---|---|---|---|
| APEX-Agents | agentic | 4.7 | 18.4 |
| Chatbot Arena Elo — Overall | arena | 1353.8 | 1467.7 |
| Chess Puzzles | knowledge | 20.0 | 12.0 |
| Fortress | safety | 8.2 | 13.6 |
| GPQA diamond | knowledge | 67.7 | 81.4 |
| OTIS Mock AIME 2024-2025 | math | 88.9 | 86.1 |
| SimpleBench | reasoning | 6.5 | 54.4 |
| SimpleQA Verified | knowledge | 13.9 | 41.8 |
| SWE-Bench Verified (Bash Only) | coding | 26.0 | 74.4 |
| Terminal Bench | coding | 18.7 | 63.1 |
| WeirdML | coding | 48.2 | 63.7 |