Medical Triage on Semigran (test)
91AccuracyPhysicians
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| PhysiciansModel=Human2026.06 | 91 | — | — | — | |
| SG-c1189Model=gpt-oss-120b2026.06 | 87.1 | 97 | 83 | 81 | |
| MedAskModel=Propr.2026.06 | 82.7 | 96 | 80 | 72 | |
| o4-miniModel=o4-mini2026.06 | 80.4 | 89 | 81 | 71 | |
| Base (best init.)Model=gpt-oss-120b2026.06 | 77.3 | 60 | 88 | 84 | |
| o3Model=o32026.06 | 75.6 | 91 | 83 | 53 | |
| o1Model=o12026.06 | 73.3 | 87 | 89 | 44 | |
| GPT-4oModel=GPT-4o2026.06 | 69.3 | 80 | 88 | 40 | |
| o3-miniModel=o3-mini2026.06 | 69.3 | 80 | 95 | 33 | |
| GPT-4.5Model=GPT-4.52026.06 | 68.9 | 93 | 83 | 31 |