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

General AI assistant tasks on GAIA n=165 (dev)

73.93Average Accuracy

OAgent

31.08242.20653.3364.454Aug 1, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.08
73.9383.0274.4253.85
2025.08
73.386.570.157.7
2025.08
70.9183.0268.653.85
2025.08
70.383.0269.7746.15
2025.08
69.0984.9167.4442.31
2025.08
67.3674.2969.0647.6
2025.08
66.6777.3666.2846.15
2025.08
63.6467.9267.4442.31
2025.08
63.6475.4761.6346.15
2025.08
63.6469.8165.1246.15
2025.08
60.6173.5862.7926.92
2025.08
58.1881.1454.6523.08
2025.08
57.5877.3654.6526.92
2025.08
56.9771.758.1423.08
2025.08
55.7671.753.4930.77
2025.08
55.1571.753.426.92
2025.08
53.9966.0452.3333.33
2025.08
52.160.3854.6526.92
2025.08
49.754.7253.4926.92
2025.08
46.0656.646.5123.08
2025.08
38.1850.9438.3711.54
2025.08
34.5447.1733.7211.54
2025.08
32.7343.432.5611.54