Results

100 questions
AgentCorrectnessResource use
Biomni0.780.76243.653.705.620.1160.148
BixBench-Agent0.760.7640.764.566.650.1280.168
BioMedAgent0.730.72167.5710.3713.210.4120.565
STELLA0.710.74367.816.2620.230.2470.347
AutoBA0.340.3972.3834.4127.100.2470.726
BioMaster0.050.1168.9214.817.770.2675.340

Performance by question

100 questions
QuestionAutoBABioMasterBiomniBixBench-AgentSTELLABioMedAgent
BL-0004FailFailFailFailFailFail
BL-0009FailFailFailFailFailFail
BL-0019FailFailPassFailFailFail
BL-0022FailFailFailFailFailFail
BL-0025FailFailFailFailFailPass
BL-0028FailFailPassFailPassPass
BL-0116PassFailPassPassPassPass
BL-0126PassFailPassPassPassPass
BL-0134PassFailPassPassPassPass
BL-0137PassFailPassPassPassPass
BL-0145PassFailPassPassPassPass
BL-0149PassPassPassPassPassPass
BL-0152PassFailPassPassPassPass
BL-0155FailFailPassPassFailPass
BL-0173FailFailPassPassPassFail
BL-0175PassFailPassPassPassPass
BL-0177PassFailPassPassPassPass
BL-0179PassFailPassPassPassPass
BL-0189PassFailPassPassPassPass
BL-0196PassFailPassPassPassPass
Task details