Medical Calculation on MedCalc-Bench Verified
34.8AccuracyDeepSeek-R1 / Gemini 2.5 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| DeepSeek-R1 / Gemini 2.5 ProApproach=Direct prompting (HELM), Setting=Zero-shot2026.02 | 34.8 |
Loading the SOTA2 catalog…
| Method | Links | |
|---|---|---|
| DeepSeek-R1 / Gemini 2.5 ProApproach=Direct prompting (HELM), Setting=Zero-shot2026.02 | 34.8 |