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

Visual Reasoning on BLINK (Granular Metrics)

99Jigsaw Accuracy

Human

36.943253.054169.16585.2759Mar 17, 2026Mar 21, 2026Mar 26, 2026Mar 31, 2026Apr 4, 2026Apr 9, 2026Apr 14, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.04
99--96.0799.42--92.48-98
2026.04
96.67--74.196.51--63.9177.6877.05
2026.03
95.885.186.763.389.484.178.9---
2026.04
91.33--64.0394.19--86.4781.4593.44
2026.04
90.67--62.5986.04--94.7369.8990.16
2026.04
88--66.1991.86--33.8361.8357.38
2026.04
86.67--64.0388.37--93.9861.0285.25
2026.04
83.33--46.0439.53--69.1732.2555.74
2026.04
80--45.3265.69--45.8640.5956.56
2026.04
79.33--60.4391.86--47.3763.7167.21
2026.04
79.33--59.7180.23--39.8441.6654.1
2026.04
78--64.0391.86--48.1257.7970.49
2026.04
76--53.2476.74--41.3552.4258.2
2026.03
75.88381.45587.376.571.6---
2026.04
75.33--33.0929.65--48.87-47.54
2026.04
75.3--58.385.5--60.2-59
2026.03
70.885.185.858.790.978.364.6---
2026.04
70.7--58.380.8--45.6-65.4
2026.04
66.67--44.664.53--75.1837.954.92
2026.04
66--53.2475.58--45.1165.0559.02
2026.04
64.67--61.8786.62--45.947.0454.92
2026.04
62--34.5341.86--56.392550
2026.04
61.33--33.8134.88--49.6229.5747.54
2026.04
60.66--62.5876.16--67.6642.445.9
2026.04
58--39.5763.95--51.8835.2148.36
2026.04
57.26--29.6148.97--44.96-60.01
2026.04
53.33--24.532.56--46.6225.2751.64
2026.04
40--38.1343.6--55.6451.0863.11
2026.04
39.33--17.9926.16--21.02-45.9