Qwen2.5 14B Instruct
Código abiertopor Alibaba · Publicado el 2024-09-16
41.6
puntuación promedio
N/A
Precio de entrada
N/A
Precio de salida
N/A
Ventana de contexto
text-generation
Tipo
Tested on 6 benchmarks with 41.6% average. Top scores: IFEval (81.4%), MATH Level 5 (55.3%), BBH (HuggingFace) (48.6%).
Puntuaciones de benchmark
| Benchmark | Categoría | Puntuación | Bar |
|---|---|---|---|
| IFEval | language | 81.4 | |
| MATH Level 5 | math | 55.3 | |
| BBH (HuggingFace) | general | 48.6 | |
| MMLU-PRO | knowledge | 43.2 | |
| MUSR | reasoning | 10.6 | |
| GPQA | knowledge | 10.5 |
Modelos similares
Mistral 7B V0.1Mistral AI
41.6
o1-previewOpenAI
41.5
Claude Opus 4Anthropic
41.7
Qwen2-72BAlibaba Qwen
41.3