Policy Evaluation on PolicyBench Overall Average
66.34AccuracyDeepseek R1
Evaluation results
| Method | Links | |
|---|---|---|
| Deepseek R12026.04 | 66.34 | |
| Claude 3.7Model Variant=Claude-3.7-Sonnet2026.04 | 64.13 | |
| Gemini 2.5Model Variant=Gemini-2.5-Flash2026.04 | 63.82 | |
| Claude 3.5Model Variant=Claude-3.5-Sonnet2026.04 | 63.75 | |
| o4-mini2026.04 | 62.97 | |
| QwQ 32B2026.04 | 61.67 | |
| Gemini 2.0Model Variant=Gemini-2.0-Flash2026.04 | 60.1 | |
| GPT-4o2026.04 | 59.47 | |
| LLaMA 42026.04 | 59.17 | |
| Deepseek V32026.04 | 59.1 | |
| Gemma 3-27B2026.04 | 58.21 |