Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Object Counting on Lens
Loading...
62.78
Accuracy
InternVL3
37.8928
44.3539
50.815
57.2761
May 21, 2025
Accuracy
Updated 20d ago
Evaluation Results
Method
Method
Links
Accuracy
InternVL3
Model size=38B
2025.05
62.78
Qwen2.5-VL
Model size=32B
2025.05
62.25
Deepseek-VL2
Model size=MoE 4.5B/27B
2025.05
61.41
InternVL3
Model size=78B
2025.05
61.38
Gemini2.5-Pro
Model size=-
2025.05
60.18
Qwen2.5-VL
Model size=72B
2025.05
59.75
Qwen2.5-VL
Model size=3B
2025.05
58.76
Qwen2.5-VL
Model size=7B
2025.05
58.35
Deepseek-VL2-tiny
Model size=MoE 1B/3B
2025.05
56.22
InternVL3
Model size=2B
2025.05
55.81
InternVL3
Model size=9B
2025.05
55.63
GPT-4o
Model size=-
2025.05
54.32
QVQ-Max
Model size=72B
2025.05
49.95
Kimi-VL-thinking
Model size=MoE 2.8B/16B
2025.05
46.87
Gemma3
Model size=12B
2025.05
44.65
Gemma3
Model size=4B
2025.05
38.85
Feedback
Search any
task
Search any
task