Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-modal Reasoning on MMK12 (test)
Loading...
50.7
Accuracy
SFT-M
41.444
43.847
46.25
48.653
Feb 11, 2026
Accuracy
Updated 3d ago
Evaluation Results
Method
Method
Links
Accuracy
SFT-M
Model=Qwen2.5-VL-7B, P...
2026.02
50.7
SFT
Model=Qwen2.5-VL-7B, P...
2026.02
49.2
GRPO
Model=Qwen2.5-VL-7B, P...
2026.02
49.05
SFT-RS
Model=Qwen2.5-VL-7B, P...
2026.02
48.6
SFT-M
Model=Qwen2.5-VL-3B, P...
2026.02
43.4
SFT
Model=Qwen2.5-VL-3B, P...
2026.02
42.2
GRPO
Model=Qwen2.5-VL-3B, P...
2026.02
42.1
SFT-RS
Model=Qwen2.5-VL-3B, P...
2026.02
41.8
Feedback
Search any
task
Search any
task