Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scientific Reasoning on Lens
Loading...
56.2
Accuracy
Gemini2.5-Pro
28.328
35.564
42.8
50.036
May 21, 2025
Accuracy
Updated 20d ago
Evaluation Results
Method
Method
Links
Accuracy
Gemini2.5-Pro
Model size=-
2025.05
56.2
Qwen2.5-VL
Model size=72B
2025.05
53.65
Qwen2.5-VL
Model size=32B
2025.05
51.66
GPT-4o
Model size=-
2025.05
51.14
QVQ-Max
Model size=72B
2025.05
50.8
InternVL3
Model size=78B
2025.05
49.39
InternVL3
Model size=38B
2025.05
47.18
Qwen2.5-VL
Model size=7B
2025.05
46.28
InternVL3
Model size=9B
2025.05
44.69
Deepseek-VL2
Model size=MoE 4.5B/27B
2025.05
44.58
Gemma3
Model size=12B
2025.05
43.33
InternVL3
Model size=2B
2025.05
40.56
Qwen2.5-VL
Model size=3B
2025.05
40.33
Gemma3
Model size=4B
2025.05
39.53
Deepseek-VL2-tiny
Model size=MoE 1B/3B
2025.05
38.97
Kimi-VL-thinking
Model size=MoE 2.8B/16B
2025.05
29.4
Feedback
Search any
task
Search any
task