Medical Knowledge Evaluation on MMLU Med
88.1AccuracyGemini 3 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 3 ProTemperature=default2026.04 | 88.1 | |
| Gemini 3 FlashTemperature=default2026.04 | 87.5 | |
| MedGemma 1Model Scale=27B, Temperature=0.02026.04 | 86.2 | |
| DOS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Distance-to-Optimum Selection2025.12 | 82.11 | |
| LPS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Low-PPL Sampling2025.12 | 81.85 | |
| HPS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=High-PPL Sampling2025.12 | 81.68 | |
| RS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Random Data Sampling2025.12 | 81.62 | |
| Qwen3-14B-BaseBackbone=Qwen3-14B-Base, CPT Strategy=None2025.12 | 81.19 | |
| Qwen3 VLModel Scale=4B, Temperature=default2026.04 | 78.3 | |
| MedGemma 1Model Scale=4B, Temperature=0.02026.04 | 70 | |
| MedGemma 1.5Model Scale=4B, Temperature=0.02026.04 | 69.7 |