Medical Calculation and Tool Use on MedMCP-Calc Special Populations & Universal Tools benchmark leaderboard · SOTA2 Research

Medical Calculation and Tool Use on MedMCP-Calc Special Populations & Universal Tools

0.6757CSCalcMate-30B

Evaluation Results

Method	Links
CalcMate-30B2026.01		0.6757	0.7438	0.1846
Claude Sonnet 4.52026.01		0.6097	0.5514	0.1472
GLM-4.72026.01		0.5743	0.4772	0.1349
Claude Opus 4.52026.01		0.5677	0.6341	0.237
CalcMate-4B2026.01		0.4994	0.6193	0.137
GPT-52026.01		0.4824	0.537	0.1733
Gemini-3-Pro2026.01		0.4671	0.4649	0.181
Gemini-2.5-Pro2026.01		0.4664	0.397	0.1908
Deepseek-V3.12026.01		0.4478	0.4729	0.1238
Kimi-K2-Instruct2026.01		0.4185	0.2468	0.0492
Qwen3-235B-Instruct-25072026.01		0.3941	0.3442	0.1126
Qwen3-235B-Thinking-25072026.01		0.3837	0.2679	0.0564
Qwen3-235B thinking2026.01		0.3371	0.3917	0.0719
Qwen3-VL-235B2026.01		0.3288	0.2552	0.1099
Qwen3-235B non-thinking2026.01		0.3243	0.3655	0.0711
Baichuan-M2-32B2026.01		0.3073	0.3346	0.0246
Qwen3-Next-80B-Instruct2026.01		0.292	0.2727	0.0383
Qwen3-30B-Thinking-25072026.01		0.2887	0.1284	0.034
Qwen3-Next-80B-Thinking2026.01		0.2802	0.2984	0.054
Qwen3-4B-Instruct-25072026.01		0.2594	0.0774	0.0194
Qwen2.5-72B2026.01		0.2546	0.3738	0.0583
Qwen3-30B-Instruct-25072026.01		0.2459	0.2677	0.0798
MedGemma-27B-IT2026.01		0.2335	0.2468	0.0308
Llama-3.3-70B2026.01		0.1375	0.2408	0.0289
Llama-3.1-70B2026.01		0.1019	0.1503	0.015