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

Information-seeking on GAIA

70.5Success Rate

OpenAI-o3

21.41234.15646.959.644Feb 11, 2026Mar 6, 2026Mar 29, 2026Apr 21, 2026May 14, 2026Jun 6, 2026Jun 30, 2026
Updated 24d ago

Evaluation Results

MethodLinks
2026.02
70.5
68.3
67.4
2026.02
65
2026.02
64.1
2026.02
61.2
2026.02
57.7
2026.02
56.3
2026.02
53.3
2026.02
53.2
2026.02
52.8
2026.02
51.5
2026.02
50
2026.06
29.1
2026.06
29.1
2026.06
25.2
2026.06
25.2
2026.06
25.2
2026.06
24.3
2026.06
23.3
2026.06
23.3
2026.06
23.3
2026.06
23.3