Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Analytical Agent Reasoning on CGM-QA synthetic queries N=2,470
Loading...
92
Precision
Gemini 3.0 Pro
43.12
55.81
68.5
81.19
Apr 18, 2026
Precision
Recall
F1 Score
Value Accuracy
Updated 3mo ago
Evaluation Results
Method
Method
Links
Precision
Recall
F1 Score
Value Accuracy
Gemini 3.0 Pro
Model Accessibility=Pr...
2026.04
92
71
80
94
Gemini 3.0 Flash
Model Accessibility=Pr...
2026.04
89
74
81
94
GPT-5.2
Model Accessibility=Pr...
2026.04
84
89
86
81
GPT-5-Mini
Model Accessibility=Pr...
2026.04
75
87
80
94
Llama-4-17B
Model Accessibility=Op...
2026.04
73
74
73
75
Nemotron-Nano-9B
Model Accessibility=Op...
2026.04
45
61
52
67
Feedback
Search any
task
Search any
task