Introducing LM Studio 0.4.0 🎉✨ It's the next generation of LM Studio. 🪄 Deploy on servers, in CI, or anywhere 🚄 Parallel requests for high throughput use cases 🔨 New stateful REST API: use local MCPs 🎨 Complete UI revamp See what's new in this release👇🧵

Jan 28, 2026 · 6:06 PM UTC

72
178
41
1,274
172,458
RelevantRecentLikes
Parallel Requests Requests to the same model can now be processed in parallel, instead of being queued. This unlocks high throughput use cases via the API, or parallel requests in the chat interface with Split View.
7
13
6
202
25,579
Daemon mode: deploy anywhere without GUI Introducing llmster. It's the core of LM Studio, packaged for the server. Combined with parallel inference requests, you can deploy llmster to power your inference serving needs anywhere.
3
2
5
106
8,956
New REST API: /v1/chat Allows you to run your local models and give them local MCPs to use. Additionally, you can now generate permission tokens to control which clients can interact with your server.
1
58
5,787
lms chat New open-source TUI for running interactive sessions with models in the terminal. Contributions are welcome!
2
1
1
73
7,539
Congrats guys, it looks awesome!
1
1
17
3,731
Thank you Maxime!
6
1,958
Love parallel requests!
2
12
1,926
🥳🦾
1
4
1,473
The new UI + local MCPs are amazing! Congrats on the launch!
1
9
3,027
API FINALLY I LOVE U GUYS TOOK U LONG ENOUGH
1
6
1,412
Congrats @yagilb and team, looks sick! 🫡
1
3
1,190
Big win! Congrats!
1
3
2,384
🏆🙌
1
1,316
This is awesome, nice work guys!! 💪🔥🚀
1
2
862
🔥🔥🔥
3
713
🚀🚀
1
1
518
🚀🚀🚀
441
Looks epic!
1
1
1,031
Thank you Karim! :)
1
632
0.4.0 の大型アップデート最高です! 0.4.1 の日本語入力まわり(lms chat の non-ASCII 修正)も手元で検証して、複数モデルで全部パスしました。 「入力が本当にモデルに届いているか」をログで確認できるの、安心感が段違いです。ありがとうございます!
1
1
138
parallel requests! great update
1
1
1,104
🚄🚄🚄
983
this can't drop on a more perfect timing for me !!
1
1
587
Holy moly
3
307
Congrats! I recently started using lm studio it’s amazing, but one thing I noticed is I can’t pin a model settings or plugin settings to a model, and have to configure every time I load a model or start a new chat. Hopefully that’s improved
3
984
Yeah - parallel requests! 🎉
2
97