What’s new: Air gets more agents, local models, and Java/Kotlin code intelligence - The JetBrains Blog

The latest release of JetBrains Air introduces several new features, including support for GitHub Copilot and other ACP-compatible agents, enhanced Java and Kotlin code intelligence, and the ability to run Windows tasks in Docker containers. Air enables developers to connect their own coding agents and harnesses, allowing tasks to be executed in parallel workspaces and reviewed before integration into the codebase. This update also allows companies to use their approved coding tools, such as GitHub Copilot, without needing additional AI providers. Local model support has been added, facilitating the use of models running offline via local runners like Ollama or LM Studio. With IntelliJ-powered navigation, users can now efficiently manage Java and Kotlin projects, identify errors, and inspect code changes directly within Air. The inclusion of Docker containers for Windows aligns with macOS capabilities, enabling isolated task execution. The update also focuses on giving users more control during agent runs, including reviewing proposed changes and monitoring context-window usage. Overall, these enhancements aim to improve the flexibility, control, and efficiency of AI-assisted coding workflows in JetBrains Air.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.