Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Medical Reasoning on MedQA-USMLE Medical (val)
Loading...
83
Accuracy+
GPT-4o
32.768
45.809
58.85
71.891
Oct 10, 2025
Accuracy+
Consistency
Exact Accuracy
Updated 25d ago
Evaluation Results
Method
Method
Links
Accuracy+
Consistency
Exact Accuracy
GPT-4o
2025.10
83
99.2
86.3
CLARITY
Model Size=7B
2025.10
63.3
89.5
66.2
Data Reformulation
2025.10
62.4
89
65.3
PRM RL
2025.10
60.2
96.4
60.8
Refine-then-Monitor
2025.10
59.7
89.3
62.3
Standard RL
2025.10
57.8
81.3
63.6
CLARITY
Model Size=3B
2025.10
50.1
86
55.1
Qwen2.5
Model Size=7B
2025.10
42.5
88.7
49.4
Qwen2.5
Model Size=3B
2025.10
34.7
77.7
38.2
Feedback
Search any
task
Search any
task