Claude 3 Opus
von Anthropic · Veroeffentlicht 2024-01-01
33.7
Durchschn. Score
N/A
Eingabepreis
N/A
Ausgabepreis
N/A
Kontextfenster
text
Typ
Tested on 8 benchmarks with 33.7% average. Top scores: MMLU (79.5%), Winogrande (77.0%), MATH level 5 (37.5%).
Benchmark-Ergebnisse
| Benchmark | Kategorie | Score | Bar |
|---|---|---|---|
| MMLU | knowledge | 79.5 | |
| Winogrande | knowledge | 77.0 | |
| MATH level 5 | math | 37.5 | |
| GPQA diamond | knowledge | 29.6 | |
| WeirdML | coding | 23.2 | |
| Cybench | coding | 10.0 | |
| SimpleBench | reasoning | 8.2 | |
| OTIS Mock AIME 2024-2025 | math | 4.6 |
Aehnliche Modelle
DeepSeek
33.6
U
XGen-7BUnknown
33.9
xAI
33.2
Google DeepMind
32.9