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

Web Reasoning on AssistantBench

31.8Accuracy

SDP

13.80818.47923.1527.821May 12, 2026
Updated 20d ago

Evaluation Results

MethodLinks
2026.05
31.83214.992.854.916
2026.05
28.3301167.848.515.5
2026.05
25.325.31169.935.616.9
2026.05
25.227.59.980.742.712.4
2026.05
23.426.19.48247.77.1
2026.05
22.224.88.367.849.74.2
2026.05
19.5216.181.3357.3
2026.05
14.520.45.563.919.38.4