{"slug": "lm-studio-is-a-free-desktop-app-to-run-local-llms-on-your-own-computer", "title": "LM Studio is a free desktop app to run local LLMs on your own computer", "summary": "LM Studio is a free desktop app that runs local large language models on macOS, Windows, and Linux, offering a private, offline alternative to cloud chatbots. The app downloads open models including gpt-oss, Llama, Qwen, DeepSeek R1, Mistral, and Gemma from Hugging Face, and serves OpenAI-compatible endpoints at localhost:1234 plus an Anthropic-compatible Messages API for Claude Code. LM Studio recommends 16GB+ RAM on Apple Silicon Macs running macOS 14.0+ and 16GB RAM with 4GB+ VRAM on x64/ARM64 Windows 10/11 and Ubuntu 20.04+ Linux systems, and also ships Python and TypeScript SDKs and the MIT-licensed lms CLI.", "body_md": "# Welcome to the LM Studio docs\n\nLM Studio is a free desktop app to **run local LLMs on your own computer** — a private, offline **ChatGPT alternative** for macOS, Windows, and Linux. Download open models like gpt-oss, Llama, Qwen, DeepSeek R1, Mistral and Gemma from Hugging Face, chat with your documents, and serve an **OpenAI-compatible local API**. To get started, head over to the [Downloads page](https://lmstudio.ai/download) and grab the installer for your operating system.\n\n## Get started with LM Studio\n\n## System requirements\n\nLM Studio generally supports Apple Silicon Macs, x64/ARM64 Windows PCs, and x64 Linux PCs.\n\n| Platform | Chip / CPU | OS | Memory | \n|---|---|---|---|\n| macOS | Apple Silicon (M1/M2/M3/M4) — Intel Macs are not supported | macOS 14.0+ | 16GB+ RAM recommended | \n| Windows | x64 (AVX2 required) or ARM (Snapdragon X Elite) | Windows 10/11 | 16GB RAM · 4GB+ VRAM recommended | \n| Linux | x64 (AVX2) or ARM64 (aarch64) | Ubuntu 20.04+ · AppImage | 16GB RAM · 4GB+ VRAM recommended | \n\n## What can I do with LM Studio?\n\n1. **Download and run local LLMs** like`gpt-oss` ,`Llama` ,`Qwen` ,`Mistral` or`DeepSeek R1` — private AI on your own hardware, no cloud account needed\n2. Use a **simple and flexible chat interface** — a private AI chatbot with system prompts, presets and streaming replies\n3. **Connect MCP servers** and use tools with local models — files, browsers, and your own automations\n4. **Search & download models via Hugging Face** 🤗 — GGUF and MLX weights, managed in one library\n5. **Serve local models on OpenAI-like endpoints** , locally and on the network — plus an Anthropic-compatible Messages API for Claude Code\n6. **Manage your local models, prompts, and configurations** — GPU offload, context length, and per-model presets\n\n## Popular use cases\n\nWhy people run LLMs locally with LM Studio instead of renting a cloud chatbot.\n\n## Run llama.cpp (GGUF) or MLX models\n\nLM Studio supports running LLMs on Mac, Windows, and Linux using `llama.cpp`. On Apple Silicon Macs, LM Studio also supports running LLMs using Apple’s **MLX** framework.\n\nTo install or manage LM Runtimes, press `⌘ Shift R` on Mac or `Ctrl Shift R` on Windows/Linux.\n\n## Run an LLM like gpt-oss, Llama, Qwen, Mistral, or DeepSeek R1 on your computer\n\nTo run an LLM on your computer you first need to download the model weights. You can do this right within LM Studio — open the Discover tab, search Hugging Face, and pick a size your machine can hold.\n\n```\n# …or from the terminal\nlms get qwen3\nlms load openai/gpt-oss-20b --gpu=max\nlms chat\n```\n\n## Offline operation\n\nLM Studio can operate entirely offline — just make sure to get some model files first. Chats, inference, and local APIs stay on your machine.\n\n## Developer — local REST API\n\nLM Studio provides a REST API that you can use to interact with local models from your own apps and scripts. Serve OpenAI-compatible endpoints (`/v1/chat/completions`, responses, embeddings) and an Anthropic-compatible Messages API — locally or on your network.\n\n```\nlms server start\ncurl http://localhost:1234/v1/models\n```\n\nFor servers, cloud instances, and CI, install **llmster** — LM Studio's headless daemon, no desktop app required:\n\n```\ncurl -fsSL https://lmstudio.ai/install.sh | bash   # macOS / Linux\nirm https://lmstudio.ai/install.ps1 | iex          # Windows\n```\n\n## Python SDK\n\nUse `lmstudio-python` to load models, generate text, embed content, and build agents from scripts, notebooks, and backend services.\n\n```\npip install lmstudio\n```\n\n## TypeScript SDK\n\nUse `lmstudio-js` to build apps, tools, and local AI workflows in Node.js or TypeScript — including plugins.\n\n```\nnpm install @lmstudio/sdk\n```\n\n## CLI — `lms`\n\n`lms` ships with LM Studio, so no extra install is needed. It is MIT licensed and developed in the open on GitHub. Bootstrap it once, then automate your local LLM workflows.\n\n```\n# Bootstrap on macOS / Linux\n~/.lmstudio/bin/lms bootstrap\n\n# Bootstrap on Windows\ncmd /c %USERPROFILE%/.lmstudio/bin/lms.exe bootstrap\n```\n\n| Command | What it does | \n|---|---|\n| `lms chat` | Chat with a model interactively in the terminal | \n| `lms get` | Search and download models | \n| `lms ls` | List models on disk (mirrors the My Models tab) | \n| `lms ps` | List models currently loaded in memory | \n| `lms load` | Load a model — options like `--gpu=max` ,`--context-length` ,`--identifier` | \n| `lms unload` | Unload a model ( `--all` for everything) | \n| `lms server start` | Control the local server | \n| `lms runtime` | Manage and update inference runtimes | \n| `lms daemon` | Manage the headless background daemon | \n| `lms link` | Manage LM Link across devices | \n\n## Bionic\n\nBionic is LM Studio’s agent for open models — natively local, built for creativity, work, and code. It excels at documents, coding tasks, automations, and real-time voice transcription, all processed on-device. For demanding tasks, run Bionic with frontier open models in a Zero Data Retention cloud.\n\n## Integrations\n\nConnect LM Studio to the tools you already use. With the Anthropic-compatible `/v1/messages` endpoint, Claude Code can run on your local models with a single base-URL change:\n\n```\nexport ANTHROPIC_BASE_URL=http://localhost:1234\nexport ANTHROPIC_AUTH_TOKEN=lmstudio\nclaude   # now running on your local model\n```\n\nAlso works with Codex, VS Code, MCP tools, and any OpenAI-compatible client.\n\n## LM Link\n\nLM Link securely connects models across your devices — route local AI workloads to preferred machines and use remote models from another device as if they were local.\n\n## Changelog\n\nCurrent release: **0.4.23**. Highlights of the 0.4.x line include the Anthropic-compatible Messages API, llmster headless daemon, and LM Link. Beta and experimental builds are published on the beta-releases page.\n\n## Frequently asked questions\n\n`lms get` and `lms chat` from the terminal. No cloud account needed.", "url": "https://wpnews.pro/news/lm-studio-is-a-free-desktop-app-to-run-local-llms-on-your-own-computer", "canonical_source": "https://www.lm-studio.net/", "published_at": "2026-09-21 13:02:36+00:00", "updated_at": "2026-09-21 13:25:37.056468+00:00", "lang": "en", "topics": ["large-language-models", "ai-tools", "ai-products", "ai-agents", "developer-tools"], "entities": ["LM Studio", "Hugging Face", "gpt-oss", "Llama", "Qwen", "DeepSeek R1", "Mistral", "Gemma"], "alternates": {"html": "https://wpnews.pro/news/lm-studio-is-a-free-desktop-app-to-run-local-llms-on-your-own-computer", "markdown": "https://wpnews.pro/news/lm-studio-is-a-free-desktop-app-to-run-local-llms-on-your-own-computer.md", "text": "https://wpnews.pro/news/lm-studio-is-a-free-desktop-app-to-run-local-llms-on-your-own-computer.txt", "jsonld": "https://wpnews.pro/news/lm-studio-is-a-free-desktop-app-to-run-local-llms-on-your-own-computer.jsonld"}}