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

Agent Security on A3S-BENCH Advanced 1.0

19.68RTR@1

Sonnet 4.5

17.359633.022348.68564.3477May 21, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.05
19.6813.778.424.35-0.21
2026.05
30.4719.8211.454.07-0.48
2026.05
35.9422.8711.7140.19
2026.05
52.5738.424.493.53-0.49
2026.05
55.0836.2420.513.55-0.62
55.2437.625.053.51-0.57
59.4946.9733.073.3-0.64
60.4742.8324.793.38-0.74
64.3247.5230.723.23-0.55
67.3753.635.533.08-0.63
77.6962.8143.682.81-0.66