Model
DeepSeek-V2
DeepSeek · open weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| BBH | 78.8 | average | N/A | Self-reported | medium | arxiv.org |
| HellaSwag | 87.1 | overall_accuracy | N/A | Self-reported | medium | arxiv.org |
| MMLU | 78.4 | exact_match | N/A | Self-reported | medium | arxiv.org |
| PIQA | 83.9 | score | N/A | Self-reported | medium | arxiv.org |
| Trivia QA | 80.0 | exact_match | N/A | Self-reported | medium | arxiv.org |
| WinoGrande | 86.3 | accuracy | N/A | Self-reported | medium | arxiv.org |
Reproduced by the community
Loading community runs…