Medical Calculation and Tool Use on MedMCP-Calc Critical Care & Perioperative Medicine
73.9CSCalcMate-4B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CalcMate-4BParameters=4 B2026.01 | 73.9 | 78.19 | 20.14 | |
| CalcMate-30BParameters=30 B2026.01 | 73.6 | 73.99 | 20.95 | |
| Claude Opus 4.5Parameters=/2026.01 | 71.3 | 73.51 | 36 | |
| Claude Sonnet 4.5Parameters=/2026.01 | 69.55 | 53.32 | 23.64 | |
| GPT-5Parameters=/2026.01 | 61.87 | 46.58 | 26.61 | |
| Gemini-2.5-ProParameters=/2026.01 | 59.61 | 41.98 | 20.83 | |
| Gemini-3-ProParameters=/2026.01 | 57.59 | 58.68 | 24.7 | |
| GLM-4.7Parameters=355 B2026.01 | 50.86 | 46.41 | 11.74 | |
| Qwen3-235B-Instruct-2507Parameters=235 B2026.01 | 48.83 | 34.95 | 11.38 | |
| Deepseek-V3.1Parameters=671 B2026.01 | 41.85 | 37.28 | 10.77 | |
| Kimi-K2-InstructParameters=1 T2026.01 | 40.54 | 26.73 | 7.5 | |
| Qwen3-Next-80B-InstructParameters=80 B2026.01 | 39.89 | 18.19 | 7.27 | |
| Qwen3-235B-Thinking-2507Parameters=235 B2026.01 | 39.74 | 28.28 | 13.77 | |
| Qwen3-Next-80B-ThinkingParameters=80 B2026.01 | 35.79 | 33.57 | 10.23 | |
| Qwen3-30B-Thinking-2507Parameters=30 B2026.01 | 34.21 | 18.92 | 6.93 | |
| Qwen3-VL-235BParameters=235B2026.01 | 32.69 | 26.13 | 7.87 | |
| Qwen3-235B non-thinkingParameters=235 B2026.01 | 32.34 | 41.55 | 8.17 | |
| MedGemma-27B-ITParameters=27 B2026.01 | 31.75 | 10.31 | 2.93 | |
| Qwen3-235B thinkingParameters=235 B2026.01 | 30.94 | 29.64 | 5.6 | |
| Qwen3-30B-Instruct-2507Parameters=30 B2026.01 | 29.99 | 23.53 | 6.07 | |
| Qwen2.5-72BParameters=72 B2026.01 | 29.93 | 41.66 | 8.04 | |
| Baichuan-M2-32BParameters=32 B2026.01 | 29.79 | 32.13 | 8.77 | |
| Qwen3-4B-Instruct-2507Parameters=4 B2026.01 | 20.7 | 15.14 | 7.47 | |
| Llama-3.1-70BParameters=70 B2026.01 | 9.22 | 16.37 | 0.73 | |
| Llama-3.3-70BParameters=70 B2026.01 | 8.04 | 24.56 | 2.4 |