Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Counting on Pixmo (test)
Loading...
70.8
Accuracy
Visual Para-Thinker
50.832
56.016
61.2
66.384
Feb 10, 2026
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
Visual Para-Thinker
Size=7B
2026.02
70.8
Qwen2.5-VL
Size=72B
2026.02
70.4
Sequential
Size=7B
2026.02
68
Qwen2.5-VL
Size=7B
2026.02
67.7
Majority voting@4
Size=7B
2026.02
67.6
Gemini2.5-Pro
Size=-
2026.02
59.8
GPT5-mini
Size=-
2026.02
54.7
GPT-4o
Size=-
2026.02
54.4
Visual Para-Thinker
Size=3B
2026.02
54.4
Claude-4-Sonnet
Size=-
2026.02
53.5
Majority voting@4
Size=3B
2026.02
52.8
Sequential
Size=3B
2026.02
52.1
Qwen2.5-VL
Size=3B
2026.02
51.6
Feedback
Search any
task
Search any
task