Medical Question Answering on MedXpert (test)
43.84AccuracyDeepseek-v3.2-685B
Evaluation results
| Method | Links | |
|---|---|---|
| Deepseek-v3.2-685BModel Group=Awesome General Models2026.01 | 43.84 | |
| Kimi-K2-Thinking-1TBModel Group=Awesome General Models2026.01 | 41.59 | |
| Gemini2.5-ProModel Group=Awesome General Models2026.01 | 39.27 | |
| DEEPMED-14B-RLModel Group=DeepResearch Models, Training Stage=RL2026.01 | 36.14 | |
| Tongyi-DeepResearch-30BA3BModel Group=DeepResearch Models2026.01 | 35.4 | |
| DEEPMED-14B-SFTModel Group=DeepResearch Models, Training Stage=SFT2026.01 | 34.8 | |
| AlphaMed-70BModel Group=Medical Reasoning Models2026.01 | 32.56 | |
| QuarkMed-32BModel Group=Medical Reasoning Models2026.01 | 28.6 | |
| HuatuoGPT-o1-70BModel Group=Medical Reasoning Models2026.01 | 26.5 | |
| Qwen3-30BA3B-ThinkingModel Group=Awesome General Models2026.01 | 26.4 | |
| M1-1K-32BModel Group=Medical Reasoning Models2026.01 | 25.2 | |
| BaiChuan-M2-32BModel Group=Medical Reasoning Models2026.01 | 22.8 | |
| Qwen3-14BModel Group=Awesome General Models2026.01 | 22.4 | |
| OursBase Model=Qwen2.5-7B-Instruct, Fine-tuning Dataset Source=m23k, Selection Method=Ours, Number of Selected Examples=20002026.06 | 17.9 | |
| RandomBase Model=Qwen2.5-7B-Instruct, Fine-tuning Dataset Source=m23k, Selection Method=Random, Number of Selected Examples=20002026.06 | 16.8 | |
| M1-1K-7BModel Group=Medical Reasoning Models2026.01 | 16.7 | |
| LearnabilityBase Model=Qwen2.5-7B-Instruct, Fine-tuning Dataset Source=m23k, Selection Method=Learnability, Number of Selected Examples=20002026.06 | 16.6 | |
| MedResaon-8BModel Group=Medical Reasoning Models2026.01 | 16.4 | |
| S2LBase Model=Qwen2.5-7B-Instruct, Fine-tuning Dataset Source=m23k, Selection Method=S2L, Number of Selected Examples=20002026.06 | 16.2 | |
| Embedding DiversityBase Model=Qwen2.5-7B-Instruct, Fine-tuning Dataset Source=m23k, Selection Method=Embedding Diversity, Number of Selected Examples=20002026.06 | 16.1 | |
| Middle PerplexityBase Model=Qwen2.5-7B-Instruct, Fine-tuning Dataset Source=m23k, Selection Method=Middle Perplexity, Number of Selected Examples=20002026.06 | 16 |