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

General AI Assistant tasks on GAIA (Pass@1, Pass@3)

70.5Pass@1 Score

OpenAI-o3

23.80435.92748.0560.173Sep 16, 2025
Updated 23d ago

Evaluation Results

MethodLinks
2025.09
70.5-
2025.09
68.3-
2025.09
63.1-
2025.09
57.7-
2025.09
51.568.9
2025.09
49.271.8
2025.09
47.363.1
2025.09
46.967
2025.09
45.661.2
2025.09
4560.2
2025.09
43.460.2
2025.09
42.461.2
2025.09
41.158.3
2025.09
40.565.1
2025.09
40.560.2
2025.09
40.156.3
2025.09
39.260.2
2025.09
35.352.4
2025.09
34.648.5
2025.09
33.352.4
2025.09
3348.5
2025.09
32.449.5
2025.09
31.744.7
2025.09
27.545.6
2025.09
27.244.7
2025.09
25.642.7