LM Studio
App : A product you use directly
A graphical interface to download and chat with open-weight models on your own PC.
The simplest entry point into local AI, provided you are willing to accept telemetry enabled by default.
Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.
What it does, concretely
Running a language model on your own computer usually requires technical knowledge: command lines, manual configuration, dependency management. LM Studio removes these barriers by offering a complete graphical interface that handles the entire process.
In practice, you open the application, browse a catalogue of open-weight models, download the one that suits you, and start chatting with it in a few clicks. The runtime engine relies on MLX (optimised for Apple chips) and llama.cpp (compatible with Windows and Linux), making it possible to leverage available hardware without additional configuration.
LM Studio also offers Bionic, an agent designed for work and code. Bionic can create and edit documents, handle programming tasks, automations, and even control your computer. It also includes real-time voice transcription processed entirely locally, with no audio data sent to any server.
This tool is aimed at curious users, professionals, and developers who want to experiment with recent models such as DeepSeek or Kimi without relying on a cloud service. However, it is not suitable for people with low-powered computers: large language models require a significant amount of RAM and, ideally, a dedicated graphics card.
What it offers
- ·Built-in catalogue of open-weight models downloadable directly from the application.
- ·Local chat interface: conversations stay on your machine, with no transit to an external server.
- ·Bionic agent for creating and editing documents, with automatic saving of changes.
- ·Bionic handles coding tasks, automations, and computer control.
- ·Real-time voice transcription processed entirely locally, supporting multiple languages.
- ·Compatibility with MLX (Apple Silicon) and llama.cpp engines to optimise performance based on available hardware.
- ·Zero Data Retention (ZDR) cloud services for features that rely on them.
- ·Free commercial use since July 2025; team plans available (paid).
In practice
- Price
- Freemium
- Commercial use
- Commercial use: yes
- Available on
- macos, windows, linux
Security and data
- Rating
- Security: one caveat, see the card
- Your data
- Data stays local
- Maintenance
- Active
Links and repository checked on 25 October 2026 · added on 1 May 2023