Ivan Kuleshov retweeted
For this release of Junie Local, we trained DFlash2 speculators for Qwen-blend-27B and Qwen3.6-27B
To train those drafters, we synthesized a diverse set of agentic trajectories solving SWE tasks
We also shipped a new inference engine based on @inco_ai Splash
open weights ↓
~40 → ~70 tokens/s on M5 Max.
Bursts up to 107. All local.
Junie Local now runs on Splash + DFlash2, with lower memory use and support for M4 Macs (64 GB RAM).
blog.jetbrains.com/junie/202…
107 tokens/second with optimized qwen3.8 on MacBook
blog.jetbrains.com/junie/202…
Ivan Kuleshov retweeted
Today we're releasing Air Governance.
For CTOs and tech leads: track AI usage, manage access, policies and spend org-wide.
Any LLM provider. Any dev tool. Truly vendor-agnostic.
Top-ups, guardrails, AI metrics, sessions.
Battle-tested at JetBrains. Can't imagine going back.
Ivan Kuleshov retweeted
Junie Local now runs on @NVIDIARTXSpark from day one.
Get a full-fledged coding agent on your PC. It reads your code, edits files, and runs tests. There are no tokens or quotas, and no code leaves the machine.
Windows support is stable, too. All you need is a 24 GB Nvidia RTX card.
Try it for free: junie.jetbrains.com/local
Ivan Kuleshov retweeted
Today we have hit a new milestone for Junie Local.
It can run on your RTX devices and Windows version is stable.
100% local.
Zero costs.
Available both though Air in IDEs / Junie CLI.
More supported hardware and more powerful models coming soon!
Now, starting at 24 GB of VRAM on Windows
junie.jetbrains.com/local
Free Junie Lite, interesting
Ivan Kuleshov retweeted
The first merch @SidePulse Pro’s with the JetBrains Junie logo.
An AI Agent status indicator for the SD Card slot.
We’ll share which conference to get them at.































