gpt-oss-120b vs Kimi K2 0711
Côte à côte. Chaque métrique. Chaque benchmark.
| Type | gpt-oss-120b | Kimi K2 0711 |
|---|---|---|
| Provider |  OpenAI |  moonshotai |
| score moyen | 46.9 | 56.2 |
| Prix d'entrée | $0.04 | $0.57 |
| Prix de sortie | $0.19 | $2.30 |
| Fenêtre de contexte | 131K tokens (~66 books) | 131K tokens (~66 books) |
| Sorti le | 2025-08-05 | 2025-07-11 |
| Code source ouvert | Open Source | Open Source |
Scores de benchmark
11 benchmarks · gpt-oss-120b: 3, Kimi K2 0711: 8
| Benchmark | Catégorie | gpt-oss-120b | Kimi K2 0711 |
|---|---|---|---|
| Aider polyglot | coding | 41.8 | 59.1 |
| Fiction.LiveBench | knowledge | 44.4 | 61.1 |
| HELM — GPQA | knowledge | 68.4 | 65.2 |
| HELM — IFEval | language | 83.6 | 85.0 |
| HELM — MMLU-Pro | knowledge | 79.5 | 81.9 |
| HELM — Omni-MATH | math | 68.8 | 65.4 |
| HELM — WildBench | reasoning | 84.5 | 86.2 |
| Lech Mazur Writing | knowledge | 77.3 | 86.9 |
| SimpleBench | reasoning | 6.5 | 11.6 |
| Terminal Bench | coding | 18.7 | 27.8 |
| WeirdML | coding | 48.2 | 39.4 |