Claude 3 Sonnet
por Anthropic 路 Publicado el 2024-01-01
28.3
puntuaci贸n promedio
N/A
Precio de entrada
N/A
Precio de salida
N/A
Ventana de contexto
text
Tipo
Tested on 6 benchmarks with 28.3% average. Top scores: MMLU (67.9%), Winogrande (50.2%), GPQA diamond (20.8%).
Puntuaciones de benchmark
| Benchmark | Categor铆a | Puntuaci贸n | Bar |
|---|---|---|---|
| MMLU | knowledge | 67.9 | |
| Winogrande | knowledge | 50.2 | |
| GPQA diamond | knowledge | 20.8 | |
| MATH level 5 | math | 18.2 | |
| WeirdML | coding | 10.2 | |
| OTIS Mock AIME 2024-2025 | math | 2.4 |
Modelos similares
Meta
28.0
OpenAI
28.6
Google DeepMind
28.0
Meta
28.7