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

Multimodal Agent Task on GAIA

47.17Success Rate (Level 1)

HF Agent

5.965216.662627.3638.0574Apr 30, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.04
47.1731.411.5433.4
2025.04
35.8516.283.8420.61
2025.04
33.9627.913.8426.06
2025.04
30.215.1017.6
2025.04
26.4211.633.8415.15
2025.04
26.4215.123.8416.97
2025.04
16.988.1409.7
2025.04
13.215.8107.27
2025.04
9.431.1603.64
2025.04
7.554.6504.85