JetBrains has released the 2024.1 updates for its IDEs with several major new features, headlined by full-line code autocompletion powered by local AI models.
The new full-line code completion functionality ensures code suggestions are processed entirely on the user’s device, minimising latency and providing a seamless offline experience.
“We’ve developed models that run directly on your device, and the IDE verifies each suggestion,” explained Mikhail Kostyukov, Product Manager in JetBrains Machine Learning Team. “This approach delivers the best user experience and ensures that the generated code never contains non-existent variables or methods.”
JetBrains’ local AI code completion provides an alternative to cloud-based offerings like GitHub’s Copilot, which has raised concerns around privacy and intellectual property. By running models directly on the user’s machine, JetBrains aims to address those issues for developers in regulated industries or security-conscious environments.
Full-line code completion comes included with existing paid JetBrains IDE subscriptions for Java, Kotlin, Python, JavaScript, TypeScript, CSS, PHP, Go, and Ruby. Support for C#, Rust, and C++ will be added in the coming months across relevant IDEs like Rider, RustRover, CLion Nova, and others.
In addition to the local AI capabilities, the 2024.1 releases introduce an overhauled terminal with visual enhancements, command completion, better navigation of nested commands, and easy access to history. Remote development also sees performance and stability improvements along with expanded technology support.
Other IDE-specific updates include Java 22 support in IntelliJ IDEA, a new K2 mode in Kotlin for improved analysis, VCS command execution using the project SDK in RubyMine, data flow analysis out of beta in GoLand, Jupyter notebook improvements in PyCharm, and Symfony AssetMapper support in PhpStorm.
The local AI models are particularly relevant for industries with strict data privacy rules like healthcare and finance, as well as developers working in protected infrastructure or with limited internet access.
While full-line completion assists with single lines, JetBrains’ AI Assistant can autocomplete entire code blocks and sees its own improvements like test generation, cloud completion, custom commit prompts, and in-editor generation.
To ensure privacy, JetBrains emphasises that its AI solutions are not trained on any customer code. The 2024.1 releases are available now.
(Photo by Natasha Connell)
See also: Python skills ‘increasingly essential’ to dev teams venturing into advanced AI
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.
Explore other upcoming enterprise technology events and webinars powered by TechForge here.