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

Visual Reasoning on HR-Bench 8K

72.6Overall Score

DeepEyes

48.99255.12161.2567.379Sep 26, 2025Oct 8, 2025Oct 20, 2025Nov 2, 2025Nov 14, 2025Nov 26, 2025Dec 9, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
72.686.858.5
2025.12
72.68758.3
2025.12
718557
2025.12
70.484.556.3
2025.09
69.9--
2025.12
69.385.550
2025.09
67.1--
2025.09
66.1--
2025.09
65.9--
2025.09
65.8--
2025.12
65.378.851.8
2025.09
61.3--
2025.09
61.1--
2025.09
60.4--
2025.09
60.1--
2025.09
60--
2025.09
59.9--
2025.09
58.3--
2025.09
57.3--
2025.09
57.3--
2025.09
56.5--
2025.12
55.56249
2025.09
54.4--
2025.09
49.9--