Accuracy on GPQA Diamond
34.3AccuracyLLaMA8B
Evaluation results
| Method | Links | |
|---|---|---|
| LLaMA8BModel=LLaMA-3.1, Router=None2025.10 | 34.3 | |
| Qwen3BModel=Qwen2.5, Router=None2025.10 | 33.3 | |
| Qwen7BModel=Qwen2.5, Router=None2025.10 | 32.8 | |
| Dr.LLMModel=Qwen2.5, Router=Enabled2025.10 | 32.8 | |
| Dr.LLMModel=Qwen2.5, Router=Enabled2025.10 | 32.4 | |
| Dr.LLMModel=LLaMA-3.1, Router=Enabled2025.10 | 32.3 | |
| Dr.LLMModel=LLaMA-3.2, Router=Enabled2025.10 | 29.8 | |
| LLaMA3BModel=LLaMA-3.2, Router=None2025.10 | 27.2 |