List your company

Medical Agent Task Execution on MedAgentBench

79.3Success RateClaude-4-Sonnet
11.59629.17346.7564.327Mar 6, 2026Mar 7, 2026Mar 8, 2026Mar 9, 2026Mar 10, 2026

Evaluation results

MethodLinks
Claude-4-SonnetCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0379.3———
Gemini-3-ProCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0374.8———
SELSMBase Model=Qwen3-30B-A3B2026.0371.3310069.3373.33
DeepSeek-V3.2-ExpCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0367.3———
Gemini-2.5-ProCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0365.1———
GPT-OSS-120BCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0364.3———
Kimi-K2-0905Category=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0361.2———
SELSMBase Model=Average2026.0360.899059.1162.66
SELSMBase Model=Qwen3-32B2026.0360.3394.3366.6654
ExpeLBase Model=Qwen3-30B-A3B2026.035899.3354.6661.33
ExpeLBase Model=Qwen3-32B2026.03589556.6659.33
DIVE-8B (RL)Category=Ours, Temperature=1, Top-p=1, Number of runs=42026.0357.3———
EnvScaler-8BCategory=8B Baselines, Temperature=1, Top-p=1, Number of runs=42026.0356.6———
SELSMBase Model=GLM4-32B2026.035175.6641.3360.66
DIVE-8B (SFT)Category=Ours, Temperature=1, Top-p=1, Number of runs=42026.0350.2———
A-MemBase Model=Qwen3-30B-A3B2026.0347.6610030.6664.66
ExpeLBase Model=Average2026.0345.5581.4444.2246.89
Qwen3-8B (base)Category=Ours, Temperature=1, Top-p=1, Number of runs=42026.0338.4———
A-MemBase Model=Average2026.0330.4492.113228.88
A-MemBase Model=Qwen3-32B2026.032489.3335.3312.66
ExpeLBase Model=GLM4-32B2026.0320.665021.3320
A-MemBase Model=GLM4-32B2026.0319.6687309.33
WebExplorer-8BCategory=8B Baselines, Temperature=1, Top-p=1, Number of runs=42026.0317.8———
SWE-Dev-8BCategory=8B Baselines, Temperature=1, Top-p=1, Number of runs=42026.0314.2———