List your company

Structured Video Reasoning on Maze Overall

90.3Exact Match PercentageTBS-20
49.42860.03970.6581.261Jun 11, 2026

Evaluation results

MethodLinks
TBS-20Budget=202026.0690.34.42
TBS-10Budget=102026.06843.53
BoN-20Budget=202026.06796.25
BoN-10Budget=102026.0675.24.01
BoN-5Budget=52026.0667.72.58
BoN-1Budget=12026.06511