Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Structured video reasoning on Maze3D Overall
Loading...
97.7
Exact Match (%)
TBS-20
85.012
88.306
91.6
94.894
Jun 11, 2026
Exact Match (%)
Normalized Time Cost
Updated 1mo ago
Evaluation Results
Method
Method
Links
Exact Match (%)
Normalized Time Cost
TBS-20
Budget=20
2026.06
97.7
1.83
TBS-10
Budget=10
2026.06
97.2
1.57
BoN-20
Budget=20
2026.06
96.7
1.95
BoN-10
Budget=10
2026.06
96.2
1.59
BoN-5
Budget=5
2026.06
95.3
1.38
BoN-1
Budget=1
2026.06
85.5
1
Feedback
Search any
task
Search any
task
