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

Claw-Eval

Benchmarks

Task NameDataset NameSOTA ResultTrend
Real-World AgentClaw-Eval
Average Score80.6
22
Personal Assistant Agent PerformanceClaw-Eval general domain 0408
Pass@370.8
13
Agent Task PerformanceClaw-Eval Continuous Mode v1 (test)
Overall Performance63.4
11
Agent Task PerformanceClaw-Eval Isolated Mode v1 (test)
Overall Success Rate64.5
11
Showing 4 of 4 rows