Qwen2 0.5B
开源来自 Alibaba · 发布于 2024-05-31
7.2
平均分
N/A
输入价格
N/A
输出价格
N/A
上下文窗口
text-generation
类型
Tested on 6 benchmarks with 7.2% average. Top scores: IFEval (18.7%), MMLU-PRO (8.0%), BBH (HuggingFace) (7.9%).
基准测试分数
| 基准测试 | 类别 | 分数 | Bar |
|---|---|---|---|
| IFEval | language | 18.7 | |
| MMLU-PRO | knowledge | 8.0 | |
| BBH (HuggingFace) | general | 7.9 | |
| MUSR | reasoning | 4.6 | |
| MATH Level 5 | math | 2.6 | |
| GPQA | knowledge | 1.4 |