Medical Diagnosis on MedEinst Baseline 1.0
69.49Baseline AccuracyECR-Agent (Qwen3-32B)
Evaluation Results
| Method | Links | |
|---|---|---|
| ECR-Agent (Qwen3-32B)Size=32B2026.01 | 69.49 | |
| GPT-5Size=-2026.01 | 54.3 | |
| Gemini-2.5-proSize=-2026.01 | 53.58 | |
| Kimi-k2Size=-2026.01 | 47.82 | |
| Baichuan-M2-32BSize=32B2026.01 | 45.03 | |
| Qwen/Qwen3-14BSize=14B2026.01 | 44.12 | |
| DeepSeek-R1Size=-2026.01 | 42.2 | |
| Claude-Sonnet-4.5Size=-2026.01 | 42.09 | |
| Qwen/QwQ-32BSize=32B2026.01 | 41.05 | |
| MedGemma-27B-text-itSize=27B2026.01 | 40.92 | |
| Qwen/Qwen3-235B-A22BSize=235B2026.01 | 40.51 | |
| Qwen/Qwen3-32BSize=32B2026.01 | 40.25 | |
| ZhipuAI/GLM-4.6Size=-2026.01 | 39.65 | |
| DyLAN (Qwen3-32B)Size=32B2026.01 | 32.11 | |
| MDAgent (Qwen3-32B)Size=32B2026.01 | 29.7 | |
| Lingshu-32BSize=32B2026.01 | 27.68 | |
| Lingshu-7BSize=7B2026.01 | 14.93 | |
| Llama3-Med42-8BSize=8B2026.01 | 6.51 |