JetBrains Rolls Out Smarter AI Edits Across IDEs

JetBrains has introduced a new feature in its AI Assistant called Next Edit Suggestions, now available in Beta for Java, Kotlin and Python across its IDEs. The update marks a shift from traditional line-by-line code completion to file-wide edit recommendations.

The feature is included in all AI Pro and AI Ultimate plans with unlimited usage and no impact on cloud quotas. Unlike standard code completion, next edit suggestions analyse recent developer edits to propose changes such as renaming variables, inserting helper methods or updating logic across the entire file. Suggestions appear in light purple near the cursor or prompt users for review if they affect other file sections.

Powered by JetBrains’ internally trained Mellum models, the system combines AI-generated edits with IDE-native actions like Rename refactoring. This integration is aimed at keeping developers in flow during tasks such as refactoring, extending logic or correcting mistakes.

For performance, the feature is supported by globally deployed models across Europe, the US and Asia, ensuring low latency similar to JetBrains’ existing code completion. 

Currently in Beta, JetBrains is working on adding more language support, deeper model tuning and expanded deterministic IDE-based code actions. Developers can enable the feature via the AI Assistant settings in supported IDEs. JetBrains is inviting users to share feedback directly within the IDE or through its blog post comments.

Recently, JetBrains unveiled Kineto, a no-code platform that generates apps and websites from prompts in under 20 minutes. Aimed at bridging creativity and technology, Kineto supports simple apps with backend, database and UI setup.