Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multimodal Evaluation on MMT-Bench
Loading...
62.65
Accuracy
Mix + CL + CARE
57.5644
58.8847
60.205
61.5253
Dec 16, 2025
Accuracy
Updated 3d ago
Evaluation Results
Method
Method
Links
Accuracy
Mix + CL + CARE
Model Size=7B
2025.12
62.65
GRPO-CARE
Model Size=7B
2025.12
62.62
Jigsaw + CL
Model Size=7B
2025.12
62.62
Jigsaw + CL + CARE
Model Size=7B
2025.12
62.52
Rotation + CL + CARE
Model Size=7B
2025.12
61.91
Jigsaw + CARE
Model Size=7B
2025.12
61.18
VisualSphinx
Model Size=7B
2025.12
60.63
Vision-Zero
Model Size=7B
2025.12
60.47
Jigsaw
Model Size=7B
2025.12
59.96
ViCrit
Model Size=7B
2025.12
59.83
Qwen 2.5 VL
Model Size=7B
2025.12
59.64
PatchFit + CL + CARE
Model Size=7B
2025.12
58.94
Visual Jigsaw
Model Size=7B
2025.12
57.76
Feedback
Search any
task
Search any
task