JunieBlog

Stories, launches, and workflow ideas from the team behind Junie.

Announcements

Junie Local Comes to the NVIDIA RTX Spark

Junie Local runs on the NVIDIA RTX Spark from day one, and Windows support is now stable. A free local AI coding agent — no tokens, no quotas, nothing leaves your machine.

Junie Lite: a free model for the work that never needed a frontier one

Junie Lite is a free model tier for Junie, the JetBrains AI coding agent. Save frontier models for hard problems, run the rest free.

Junie Local: Smarter, Faster AI Coding

Junie Local now ships Qwen3.8-3.6-27B-blend, a merged model that completes more coding tasks than Qwen3.6 while generating 71% fewer output tokens than Qwen3.8 – plus experimental NVIDIA RTX support on Windows.

Your diff has a demo now with Junie /demo

Let Junie test a feature in your app, watch it work, and review the video, screenshots, and HTML report.

Get Gemini 3.8 Flash for Junie, Now With 75% Off

Gemini 3.8 Flash, Google's newest coding model, is tuned for long jobs and now available in Junie's IDE plugin and CLI, with a limited-time introductory offer of 75% off.

Junie Can Now Run Entirely on Your Mac – No Credits, No Cloud

Introducing Junie Local: a coding agent that runs entirely on your Mac with a model we chose and tuned against our own agent loop – no tokens, no quota, no code leaving your machine, and free. Install it with a single /local command.

Junie's New Default Runs on Gemini 3.7 Flash, at 70% Off Base Pricing

Gemini 3.7 Flash is Junie's new default model in the IDE plugin and Junie CLI: Google's most capable Flash model for coding, at 70% off base pricing for a limited time.

Junie: The JetBrains AI Coding Agent Leaves Beta

Junie is now generally available: the JetBrains AI coding agent plans before it codes, debugs with the real debugger, reviews PRs, and runs tasks async.

Keep in the loop with the Junie newsletter