The introduction of the 3.5 Sonnet model marked a significant improvement in programming capabilities, enabling Cursor to enhance its features and user experience.

Every significant feature in Cursor now involves a custom model, which was a surprising and counterintuitive development for us, and Cursor's autocomplete experience predicts the next actions across multiple files and locations within a codebase, requiring models to be fast and specialized in understanding code diffs.

The new models, Claude Opus 4 and Claude Sonnet 4, have impressed us, particularly with their capabilities in agentic coding, which we are excited to integrate into Cursor.

The progression of AI models has allowed Cursor to continuously improve its features, enabling more effective coding and problem-solving.

Our updated 2.5 Pro model has surpassed the previous version by 140 Elo points and is now the fastest growing model on Cursor, producing hundreds of thousands of lines of accepted code every minute.

The ensemble of models approach allows Cursor to leverage the strengths of various models, optimizing for both cost and performance in code generation.

The model's vision of how people would program finally became effective, leading to competition with Cursor.

Building models for Cursor involves training them to autocomplete changes in a codebase, focusing on speed and accuracy in predicting both deletions and additions.

Cursor and other AI-native integrated development environment tools are rapidly emerging, with 17% of respondents already using Cursor regularly.