Top Models
Full leaderboard →
GPT-5.2
Best
Overall Accuracy
66.4%
Resolution
40.1%
Monetary
53.1%
Injunction
79.6%
Officer Bar
92.8%
GPT-4o
Overall Accuracy
64.9%
Resolution
38.6%
Monetary
53.0%
Injunction
78.8%
Officer Bar
89.2%
Gemini 2.0
Overall Accuracy
62.8%
Resolution
38.9%
Monetary
46.9%
Injunction
79.6%
Officer Bar
85.7%
Claude Opus 4
Overall Accuracy
46.8%
Resolution
32.4%
Monetary
41.2%
Injunction
68.4%
Officer Bar
75.6%
Higher score = better. Rankings based on accuracy predicting real SEC enforcement outcomes.
More models coming soon
