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

Web Agent Task Completion on Browsecomp

51.5Accuracy

OpenAI DeepResearch

0.95614.07827.240.322Jun 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
51.5--
2026.06
49.7--
2026.06
44.747.6311,093
2026.06
42.861.2811,522
4163.712,014
2026.06
38.860.0710,894
2026.06
37.362.812,222
2026.06
32.847.2210,883
2026.06
30.965.5312,185
2026.06
13.729.436,289
2026.06
12.2--
2026.06
9.416.653,426
2026.06
4.613.71,837
2026.06
3.84.781,317
2026.06
2.925.81,683