Llama 3.2 3B Instruct (free)
Open Sourcedi Meta · Rilascio 2024-09-25
8.7
punteggio medio
N/A
Prezzo Input
N/A
Prezzo Output
131K tokens (~66 books)
Finestra di Contesto
text
Tipo
Tested on 6 benchmarks with 8.7% average. Top scores: MMLU-PRO (16.5%), BBH (HuggingFace) (14.2%), IFEval (13.4%).
Punteggi Benchmark
| Benchmark | Categoria | Punteggio | Bar |
|---|---|---|---|
| MMLU-PRO | knowledge | 16.5 | |
| BBH (HuggingFace) | general | 14.2 | |
| IFEval | language | 13.4 | |
| MUSR | reasoning | 3.8 | |
| GPQA | knowledge | 2.4 | |
| MATH Level 5 | math | 1.9 |
Modelli Simili
Alibaba
10.1
Alibaba
7.2
Google DeepMind
10.4
DeepSeek
10.4