Qwen/Qwen2-7B
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| HellaSwag | knowledge | 83.2 | HellaSwag | medium | | HellaSwag | **83.2** | 82.2 | 82.1 | 78.5 | 80.7 | |
| WinoGrande | knowledge | 78.4 | Winogrande | medium | | Winogrande | 78.4 | **79.0** | 77.4 | 71.3 | 77.0 | |
| MMLU | knowledge | 64.2 | MMLU | medium | | MMLU | 64.2 | 64.6 | 66.6 | 61.0 | **70.3** | |
| BBH | reasoning | 56.1 | BBH | medium | | BBH | 56.1 | 55.1 | 57.7 | 40.2 | **62.6** | |
| GSM8K | math | 52.2 | GSM8K | medium | | GSM8K | 52.2 | 46.4 | 56.0 | 62.5 | **79.9** | |
| MBPP | coding | 51.1 | MBPP | medium | | MBPP | 51.1 | 50.6 | 53.9 | 51.6 | **65.9** | |
| CMMLU | knowledge | 50.8 | CMMLU | medium | | CMMLU | - | - | 50.8 | 73.1 | **83.9** | |
| C-Eval | knowledge | 47.4 | C-Eval | medium | | C-Eval | 47.4 | 43.6 | 49.5 | 74.1 | **83.2** | |
| TruthfulQA | safety | 42.2 | TruthfulQA | medium | | TruthfulQA | 42.2 | 44.8 | 44.0 | 51.1 | **54.2** | |
| EvalPlus | coding | 36.4 | EvalPlus | medium | | EvalPlus | 36.4 | 39.6 | 40.3 | 40.0 | **54.2** | |
| MMLU-Pro | knowledge | 30.9 | MMLU-Pro | medium | | MMLU-Pro | 30.9 | 33.7 | 35.4 | 29.9 | **40.0** | |
| HumanEval | coding | 29.3 | HumanEval | medium | | HumanEval | 29.3 | 37.2 | 33.5 | 36.0 | **51.2** | |
| GPQA | reasoning | 24.7 | GPQA | medium | | GPQA | 24.7 | 25.7 | 25.8 | 26.7 | **31.8** | |
| MATH | math | 13.1 | MATH | medium | | MATH | 13.1 | 24.3 | 20.5 | 20.3 | **44.2** | |
none scraped.