Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multimodal Reasoning Efficiency on MM-Vet
Loading...
3.73
Time (s/sample)
MIRROR
1.3144
17.6197
33.925
50.2303
Feb 21, 2026
Time (s/sample)
Tokens (per sample)
Updated 4d ago
Evaluation Results
Method
Method
Links
Time (s/sample)
Tokens (per sample)
MIRROR
2026.02
3.73
112.58
Adaptive-CoF
Paradigm=Thinking with...
2026.02
4.02
143.51
DeepEyes
Paradigm=Thinking with...
2026.02
5.06
197.51
VL-Rethinker
Paradigm=Text Reflection
2026.02
5.56
229.33
PixelReasoner
Paradigm=Thinking with...
2026.02
64.12
273.95
Feedback
Search any
task
Search any
task