Kilo Code
The open-source agentic engineering platform
5.0•46 reviews•5.8K followers
The open-source agentic engineering platform
5.0•46 reviews•5.8K followers
Build, ship, and iterate faster with the most popular open-source AI coding agent. 5M+ Kilo Coders. 10T+ tokens processed per month.
Follow us: kilo.ai/social
This is the 5th launch from Kilo Code. View more
Kilo Code for JetBrains
Launching today
Kilo Code for JetBrains is a fully native, open-source coding agent for IntelliJ IDEA, WebStorm, PyCharm, GoLand, Rider, PhpStorm, CLion, RubyMine — any JetBrains IDE. Designed for both local and remote dev, with parallel agents in isolated worktrees, GitHub PRs and diffs inline, and 500+ models.
Launch Team / Built With
Kilo Code
@Kilo Code is so back on @Product Hunt.
Four months ago, the team launched a new @VS Code extension. #1 Product of the Month, the community loved it.
Today, we’re shipping a native plugin for @JetBrains. Rebuilt from the ground up in Kotlin with Swift UI, designed for both local and remote dev, the plugin brings first-class AI coding assistance to IntelliJ IDEA, WebStorm, PyCharm, and other JetBrains IDEs: parallel agents, GitHub PRs and diffs inline, 500+ models...
View source code on GitHub: github.com/kilo-org/kilocode
Looking forward to seeing what you're building with @Kilo Code.
Great tool, thank you! It's a hassle to force AI agents to use JetBrains instead of their primitive tools.
Does it support Claude via subscription?
Kilo Code
@artk yes, you can BYOK with Kilo
The native JetBrains support is a big plus. Does it feel as smooth as using Kilo in VS code?
Kilo Code
it is! the new @Kilo Code for @JetBrains is designed for both local and remote dev, with parallel agents in isolated worktrees, GitHub PRs and diffs inline, and 500+ models. no brainer.
How well does it handle agents needing different models for different tasks? For example, one model for planning and another for implementation could be useful.
Kilo Code
@jharkhandi_chora1 yes! you can do different things: you can assign a default model or a mode (like "Plan") and/or confgure a subagent: write a prompt that would help the agent choose that and choose mode/effort;
Settings -> Kilo Code -> Models and Settings -> Kilo Code -> Agent Behavior -> Agents
This looks like a solid step for Ai-assisted coding, How does kilo code handle large and more complex codebase?
Does the JetBrains version support custom model configurations? I could see experienced developers waiting pretty fine grained control over that.
Having 500+ models sounds useful, but i am curious how easy it is to pick the right one for a specific task.