Full Trajectory Prediction on EHRWorld-110K (test)
47.5S@10EHRWorld-14B
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| EHRWorld-14BSize=14B2026.02 | 47.5 | 58.2 | 71.6 | 26.2 | 66.7 | 92.5 | 90.1 | 91.3 | 76.5 | |
| EHRWorld-8BSize=8B2026.02 | 46.8 | 57.7 | 71.4 | 26.9 | 65.8 | 93.6 | 89.1 | 91.3 | 76.2 | |
| EHRWorld-4BSize=4B2026.02 | 46 | 56.6 | 70.3 | 27.4 | 64.9 | 93.9 | 88.6 | 91.2 | 75.5 | |
| Gemini-3.0-Pro-PreviewSize=-2026.02 | 44.5 | 54.3 | 68.1 | 29.9 | 56.5 | 48.1 | 47.3 | 47.7 | 57.4 | |
| GLM-4.7Size=358B2026.02 | 44.2 | 52.5 | 66.3 | 32.2 | 47.6 | 38.1 | 25.4 | 30.5 | 48.1 | |
| GPT-5.2Size=-2026.02 | 43.5 | 54 | 67.4 | 34.8 | 62.7 | 63.5 | 49 | 55.3 | 61.8 | |
| DeepSeek-V3.2Size=671B2026.02 | 41.6 | 51.8 | 64.9 | 34 | 55.4 | 50.9 | 40.2 | 44.9 | 55.1 | |
| MiniMax-M2.1Size=229B2026.02 | 41 | 51.3 | 65 | 33.8 | 50.2 | 47.5 | 48.6 | 48 | 54.4 | |
| Qwen3-235B-A22B-InstructSize=235B2026.02 | 40.9 | 51.4 | 64.7 | 33.9 | 46.9 | 48.3 | 34.3 | 40.2 | 50.6 | |
| Qwen3-30B-A3B-InstructSize=30B2026.02 | 40.3 | 50.8 | 64.5 | 33.8 | 41.9 | 32.4 | 41.5 | 36.4 | 47.6 | |
| GPT-OSS-120BSize=120B2026.02 | 40.2 | 50.2 | 63.4 | 37.2 | 48.3 | 43.2 | 34.6 | 38.4 | 50 | |
| Qwen3-Next-80B-A3B-InstructSize=80B2026.02 | 39.8 | 50.1 | 63 | 38.1 | 53.8 | 41.6 | 47.1 | 44.2 | 53.7 | |
| Qwen2.5-14B-InstructSize=14B2026.02 | 39.3 | 50.7 | 65.9 | 33.9 | 42.3 | 35.8 | 23.7 | 28.5 | 45.6 | |
| Llama-3.3-70B-InstructSize=70B2026.02 | 38.1 | 48.4 | 62.5 | 36.4 | 37.7 | 58.1 | 43.9 | 50 | 50.1 | |
| Baichuan-M2-32BSize=32B2026.02 | 37.2 | 47.2 | 60.1 | 41.8 | 49.9 | 45.6 | 38.6 | 41.8 | 50.6 | |
| Qwen3-4B-InstructSize=4B2026.02 | 36.5 | 47.1 | 60.7 | 44.8 | 47.3 | 45 | 25.8 | 32.8 | 46.9 | |
| MedGemma-27B-ITSize=27B2026.02 | 36.4 | 46 | 58.8 | 49.6 | 43.6 | 25.9 | 20.5 | 22.9 | 41.8 | |
| Qwen3-8BSize=8B2026.02 | 35.9 | 46.3 | 60 | 43.7 | 46.6 | 44.8 | 37.3 | 40.7 | 49.1 | |
| MedGemma-4B-ITSize=4B2026.02 | 33.6 | 43 | 55.4 | 56.1 | 43.8 | 41.2 | 14.2 | 21.1 | 40.1 |