Tool Use on Multi-Calculator
59.97R-F1MedCalc-Pro Agent Framework
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MedCalc-Pro Agent FrameworkModel=gpt-oss-120b2026.07 | 59.97 | 53.87 | 59.53 | |
| MedCalc-Pro Agent FrameworkModel=GPT-5-mini2026.07 | 57.73 | 44.7 | 48.94 | |
| MedCalc-Pro Agent FrameworkModel=Qwen3-235B-A22B2026.07 | 54.44 | 45.58 | 44.37 | |
| ReActModel=gpt-oss-120b2026.07 | 54.32 | 34.8 | 37.57 | |
| MedCalc-Pro Agent FrameworkModel=Llama-3.3-70B-Instruct2026.07 | 53.53 | 44.53 | 46.15 | |
| MedCalc-Pro Agent FrameworkModel=DeepSeek-V3.12026.07 | 52.63 | 39.65 | 43.18 | |
| ReActModel=GPT-5-mini2026.07 | 50.79 | 31.99 | 35.12 | |
| MedCalc-Pro Agent FrameworkModel=MedResearcher-R1-32B2026.07 | 50.33 | 40.14 | 37.5 | |
| ReActModel=Qwen3-235B-A22B2026.07 | 50.21 | 30.38 | 33.59 | |
| ReActModel=DeepSeek-V3.12026.07 | 48.58 | 28.91 | 30.84 | |
| ReActModel=Llama-3.3-70B-Instruct2026.07 | 47.01 | 25.21 | 27.2 | |
| ReActModel=MedResearcher-R1-32B2026.07 | 44.99 | 31.82 | 34.41 | |
| MeNTiModel=DeepSeek-V3.12026.07 | 34.99 | 23.17 | 21.11 | |
| MeNTiModel=Llama-3.3-70B-Instruct2026.07 | 33.83 | 21.11 | 26.68 | |
| MeNTiModel=GPT-5-mini2026.07 | 32.38 | 24.62 | 28.28 | |
| MeNTiModel=Qwen3-235B-A22B2026.07 | 31.95 | 19.26 | 17.47 | |
| MeNTiModel=MedResearcher-R1-32B2026.07 | 29.55 | 17.73 | 15.79 | |
| MeNTiModel=gpt-oss-120b2026.07 | 29.06 | 15.12 | 15.61 | |
| MedRaCModel=DeepSeek-V3.12026.07 | 21.12 | 6.86 | 12.89 | |
| MedRaCModel=Llama-3.3-70B-Instruct2026.07 | 20.72 | 6.22 | 14.49 | |
| MedRaCModel=gpt-oss-120b2026.07 | 20.38 | 5.29 | 13.71 | |
| MedRaCModel=MedResearcher-R1-32B2026.07 | 20.38 | 6.33 | 12.78 | |
| MedRaCModel=Qwen3-235B-A22B2026.07 | 18.1 | 5.53 | 11.26 | |
| MedRaCModel=GPT-5-mini2026.07 | 18 | 6.97 | 12.3 |