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

Scientific Agent Task Completion on ScienceAgentBench

43.1Success Rate (SR)

Mimosa

2.22812.83923.4534.061Mar 30, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.03
43.192.11.7
2026.03
38.289.80.05
2026.03
32.479.40.38
2026.03
31.377.32.2
2026.03
30.388.57.8
2026.03
21.672.13.5
2026.03
18.6641
2026.03
13.568.11.3
2026.03
7.8571.46
2026.03
3.860.70.56