Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Scientific Code Generation on ScienceAgentBench

25.5SR

SciNav

6.7811.6416.521.36Mar 11, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.03
25.572.50.893
2026.03
22.584.30.066
2026.03
19.667.60.298
2026.03
18.669.90.342
2026.03
18.659.80.023
2026.03
16.1660.512
2026.03
15670.03
2026.03
14.771.20.057
2026.03
13.162.81.093
2026.03
7.542.20.011