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

Multimodal Logical Reasoning on LogicVista (test)

64.4Score

GPT-4o

30.70439.45248.256.948Mar 18, 2026
Updated 29d ago

Evaluation Results

MethodLinks
2026.03
64.4
2026.03
54.8
52.9
51.7
2026.03
50
2026.03
48.9
2026.03
48.5
2026.03
47.9
2026.03
46.3
2026.03
45.8
2026.03
44.5
2026.03
43.6
2026.03
42.7
2026.03
42.7
2026.03
41.4
2026.03
32