Llama 3.1 70B Instruct vs Llama 3.3 70B Instruct
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | Llama 3.1 70B Instruct | Llama 3.3 70B Instruct |
|---|---|---|
| Provider |  Meta | Meta |
| puntuación promedio | 37.8 | 46.9 |
| Precio de entrada | $0.40 | $0.10 |
| Precio de salida | $0.40 | $0.32 |
| Ventana de contexto | 131K tokens (~66 books) | 131K tokens (~66 books) |
| Publicado el | 2024-07-23 | 2024-12-06 |
| Código abierto | Open Source | Open Source |
Puntuaciones de benchmark
8 benchmarks · Llama 3.1 70B Instruct: 2, Llama 3.3 70B Instruct: 6
| Benchmark | Categoría | Llama 3.1 70B Instruct | Llama 3.3 70B Instruct |
|---|---|---|---|
| Aider — Code Editing | coding | 58.6 | 59.4 |
| BBH (HuggingFace) | general | 55.9 | 56.6 |
| Chatbot Arena Elo — Overall | arena | 1292.8 | 1318.0 |
| GPQA | knowledge | 14.2 | 10.5 |
| IFEval | language | 86.7 | 90.0 |
| MATH Level 5 | math | 38.1 | 48.3 |
| MMLU-PRO | knowledge | 47.9 | 48.1 |
| MUSR | reasoning | 17.7 | 15.6 |