Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scientific Multimodal Reasoning on MMK-12
Loading...
58.5
Physics Accuracy
PTA-GRPO
44.876
48.413
51.95
55.487
Oct 2, 2025
Physics Accuracy
Chemistry Accuracy
Biology Accuracy
Updated 7d ago
Evaluation Results
Method
Method
Links
Physics Accuracy
Chemistry Accuracy
Biology Accuracy
PTA-GRPO
Base Model=Qwen2.5-7B-VL
2025.10
58.5
68.7
66.8
SRPO
Base Model=Qwen2.5-7B-VL
2025.10
56.2
65.2
65.2
Base
Base Model=Qwen2.5-7B-VL
2025.10
45.4
56.4
54
Feedback
Search any
task
Search any
task
