List your company

GUI Task Automation on OS-Marathon Transcript domain (Level 2)

86.1SWAHuman
-3.44419.80343.0566.297Jan 28, 2026

Evaluation results

MethodLinks
HumanStep=2002026.0186.150
HumanStep=1502026.0169.1825
HumanStep=1002026.0153.0425
AgentS2.5 + GPT5 w/ FCWDStep=2002026.0142.050
AgentS2.5 + GPT5 w/ FCWDStep=1502026.0138.780
HumanStep=502026.0127.380
AgentS2.5 + GPT5 w/ FCWDStep=1002026.0125.080
AgentS2.5 + GPT5Step=2002026.0123.530
AgentS2.5 + GPT5Step=1502026.0122.060
AgentS2.5 + GPT5Step=1002026.0117.650
AgentS2.5 + GPT5 w/ FCWDStep=502026.0110.610
AgentS2.5 + GPT5Step=502026.015.880
OpenCUA-7BStep=502026.0100
OpenCUA-7BStep=1002026.0100
OpenCUA-7BStep=1502026.0100
OpenCUA-7BStep=2002026.0100
UI-TARS-1.5-7BStep=502026.0100
UI-TARS-1.5-7BStep=1002026.0100
UI-TARS-1.5-7BStep=1502026.0100
UI-TARS-1.5-7BStep=2002026.0100