How to Choose the Right AI Model in Cursor?
Choose the best AI model in Cursor! Claude 3.7 vs. Gemini 2.5 Pro – expert tips for devs. Optimize speed, cost & performance.
"AI Disruption" Publication 6300 Subscriptions 20% Discount Offer Link.
Recently, Cursor released a ranking of the most popular AI models among developers.
Official data reflects the choices of the majority, with Claude 3.7 Sonnet firmly securing the top spot. So, does that mean you should just blindly choose Claude 3.7 Sonnet?
Not necessarily!
Take me as an example—my usage ratios differ significantly from the official rankings.
In Cursor, Gemini 2.5 Pro accounts for 80%, Claude 3.7 Sonnet 10%, GPT-3.5 5%, and finally GPT-4.1 5%.
In Roo and Cline, Grok dominates with 90%, while Gemini 2.5 Flash takes up 10%, and other models are barely used.
What are the considerations or basic principles here? Below, I’ll share some of my personal preferences: