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

Deep Research on DeepResearch Bench (test)

56.28Comprehensiveness

VeriTrace

44.382447.471250.5653.6488May 25, 2026
Updated 7d ago

Evaluation Results

MethodLinks
2026.05
56.2859.5652.2451.9855.77235.6673.28
2026.05
54.2555.8552.4751.5453.94139.9176.17
2026.05
52.8457.0851.6549.6153.5--
2026.05
51.295149.9848.8950.62166.7378.25
2026.05
50.8555.2250.3849.0851.96--
2026.05
50.8455.8650.7249.2952.28134.268.46
2026.05
50.5151.4551.725050.92--
2026.05
50.0650.7651.3149.7250.622.4434.74
2026.05
49.751.2450.5250.6150.62-72.5
2026.05
49.5149.4550.125049.71165.3478.3
2026.05
46.4643.7349.3947.2246.4539.7975.01
2026.05
45.3442.7947.5844.6645--
2026.05
44.9641.9747.1445.5944.64--
2026.05
44.8440.5647.9544.6944.3452.6252.86