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

Video Reasoning on TUNA-Bench

66.2Accuracy

OmniJigsaw (CMM)

48.478453.079257.6862.2808Apr 9, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.04
66.2
2026.04
65.29
2026.04
65.15
2026.04
64.73
2026.04
64.39
2026.04
63.9
2026.04
63.62
2026.04
63.41
2026.04
62.78
2026.04
62.57
2026.04
55.94
2026.04
53.84
2026.04
52.65
2026.04
49.44
2026.04
49.16