Qwen2 1.5B Instruct
Open Sourcevon Alibaba · Veroeffentlicht 2024-06-03
14.1
Durchschn. Score
N/A
Eingabepreis
N/A
Ausgabepreis
N/A
Kontextfenster
text-generation
Typ
Tested on 6 benchmarks with 14.1% average. Top scores: IFEval (33.7%), MMLU-PRO (16.7%), BBH (HuggingFace) (13.7%).
Benchmark-Ergebnisse
| Benchmark | Kategorie | Score | Bar |
|---|---|---|---|
| IFEval | language | 33.7 | |
| MMLU-PRO | knowledge | 16.7 | |
| BBH (HuggingFace) | general | 13.7 | |
| MUSR | reasoning | 12.0 | |
| MATH Level 5 | math | 7.2 | |
| GPQA | knowledge | 1.6 |
Aehnliche Modelle
Alibaba
14.3
Meta
14.5
Alibaba Qwen
13.5
HF
SmolLM2 135M InstructHugging Face TB
15.0