Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Medical Agentic Reasoning on MedAgentBench
Loading...
87
Accuracy
Static Workflow
-3.48
20.01
43.5
66.99
May 29, 2026
Jun 2, 2026
Jun 6, 2026
Jun 10, 2026
Jun 14, 2026
Jun 18, 2026
Jun 23, 2026
Accuracy
Cost (USD / 100 examples)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Cost (USD / 100 examples)
Static Workflow
Workflow Type=Static,...
2026.05
87
0.64
Dynamic Workflow (ReAct)
Workflow Type=Dynamic,...
2026.05
69
0.65
Nemotron-Terminal-32B
Train Size=264K, Metho...
2026.06
62.6
-
OpenThinkerAgent-32B
Train Size=100K, Metho...
2026.06
47.8
-
SWE-Lego-Qwen3-32B
Train Size=18K, Method...
2026.06
36.2
-
SA-SWE-32B
Train Size=4.5K, Metho...
2026.06
15.8
-
SERA-32B
Train Size=25K, Method...
2026.06
15.6
-
DeepSWE-Preview
Train Size=4.5K, Metho...
2026.06
8.7
-
Qwen3-32B
Train Size=N/A, Method...
2026.06
6.8
-
Zero-shot (Default Imp.)
Workflow Type=Zero-sho...
2026.05
0
0.22
Feedback
Search any
task
Search any
task