# LM Studio Published January 1, 2026 · Updated August 10, 2026 Discover, download, and run local LLMs with a desktop GUI Category: Developer Tools · Free · Replaces Private LLM ($4.99) · [Official website](https://lmstudio.ai) ## Install with Homebrew `brew install --cask lm-studio` ## Quick take In August 2026 LM Studio is the default GUI for serious local LLM work on Mac, now paired with Bionic for agentic coding and docs. Free-for-work licensing, MLX performance, and a rapid release train make it hard to beat if you want open models under your control. Best for: Mac users running local LLMs, Developers wiring OpenAI-compatible tools to localhost, Privacy-sensitive teams avoiding cloud prompts ## What is LM Studio? LM Studio is a free desktop app for finding, downloading, and running open large language models on your own machine. On macOS it is especially strong on Apple Silicon: a first-class MLX engine sits beside llama.cpp/GGUF support, an OpenAI-compatible local server, developer SDKs, and MCP (Model Context Protocol) hosting. You chat offline, expose models to other tools on localhost, and keep prompts on-device unless you turn on cloud features on purpose. Since May 2026 the product line grew quickly. LM Studio 0.4.0 (28 January 2026) added server-oriented deployment, parallel requests with continuous batching, a refreshed UI, and a stronger REST surface; later 0.4.x builds improved MLX batching for multi-model routing. Homebrew currently tracks 0.4.12 as a stable install path (`brew install --cask lm-studio`). In April 2026 Locally AI joined LM Studio to deepen Apple platform work, and June 2026 shipped LM Link on iPhone/iPad via the Locally app so large models on a Mac can be used from mobile. The headline launch is LM Studio Bionic (16 July 2026): a separate agent app for coding, documents, slides, and voice, powered by the LM Studio runtime (MLX + llama.cpp), local models, LM Link, or frontier open models on LM Studio Secure Cloud with Zero Data Retention. Commercial use of LM Studio itself has been free for work since July 2025, with no separate commercial licence form required. ## Features - **Local model browser & chat.** Search Hugging Face-compatible catalogs, download GGUF and MLX builds, and chat offline. Resource estimates help you pick quantisations that fit unified memory on M-series Macs. Speculative decoding, idle TTL, and runtime auto-updates improve day-to-day use. - **Apple MLX engine.** Native MLX inference on Apple Silicon delivers strong tokens-per-second and efficient memory use versus CPU-only paths. Unified multi-modal MLX work landed earlier; 2026 releases focused on agentic workloads, KV-cache checkpointing (mlx-engine improvements), continuous batching, and M5 fixes. - **OpenAI-compatible server & developer APIs.** Turn on a local server for /v1/chat/completions and /v1/responses-style workflows. Python and JavaScript SDKs, the `lms` CLI, Anthropic-compatible endpoints for tools like Claude Code, and Open Responses support let agents and IDEs call local models without rewriting clients. Default listen port is localhost:1234. - **MCP host & tool calling.** LM Studio can host Model Context Protocol servers so local models use tools, files, and outside integrations. Tool-calling reliability improved across 0.3.x-0.4.x for families such as gpt-oss, GLM, Qwen, and Ministral variants as they ship. - **LM Studio Bionic agent.** Bionic is a separate app launched July 2026 for agentic coding and knowledge work: project folders with inline diffs, sandboxed document/slide/spreadsheet work, automatic checkpoints, local voice transcription (Voxtral at launch), and optional Secure Cloud open models (for example Kimi K3, DeepSeek V4 family, GLM 5.2) under Zero Data Retention. - **LM Link & Locally mobile.** After Locally AI joined LM Studio (April 2026), LM Link reached iPhone and iPad so you can drive large models running on a Mac from mobile clients. Handy for Mac Studio or Mac mini boxes left online as local servers. - **Free for personal and work use.** The desktop app is free to download. Since July 2025 LM Studio is explicitly free for use at work without commercial licence paperwork. Cloud inference in Bionic is optional metered usage when you choose Secure Cloud models. ## How to Install LM Studio on Mac LM Studio supports macOS on Apple Silicon (recommended) and Intel. Download the latest macOS build from lmstudio.ai. For agent workflows also install LM Studio Bionic. Grant any requested Local Network / Automation permissions if you use LM Link or external tool bridges. 1. **Install via Homebrew.** Run `brew install --cask lm-studio` in your terminal. This downloads and installs the latest stable version (0.4.12) of LM Studio to your Applications folder. 2. **Launch and Browse Models.** Open LM Studio from your Applications folder. The home screen shows featured models and a search bar. Browse the model library and select a model that fits your RAM (shown in the listing). 3. **Download a Model.** Click download next to your chosen model. For first-time users, try Llama 3.1 8B Q4_K_M (about 4.7GB). It runs well on 16GB Macs and handles most general tasks. 4. **Start Chatting or Serving.** Switch to the Chat tab to start a conversation, or go to the Server tab and click 'Start Server' to expose the OpenAI-compatible API on localhost:1234. ## Pros - Strong GUI for local LLMs on Apple Silicon with MLX and GGUF - Free for work with no commercial licence friction - OpenAI-compatible server, SDKs, MCP, and Claude Code-friendly endpoints - Bionic adds coding and docs workflows without abandoning open models - Active 2026 release cadence (0.4.x, MLX work, mobile LM Link) - Local-by-default design with an optional ZDR cloud path ## Cons - Large models still need 32-64 GB+ unified memory for comfortable speeds - Bionic is a separate app to install and learn beside classic LM Studio - Cloud open-model usage is optional but metered once you leave pure local mode - GUI overhead means headless Ollama/MLX-LM can be faster for pure automation - Model quality varies widely; quant choice still matters ## Where LM Studio sits in the local AI stack LM Studio is the GUI-first path into local models on Mac, while Ollama and mlx-serve own more of the headless automation niche. That split matters for adoption: people who bounce off llama.cpp flags still finish a first chat here. On Apple Silicon, prefer MLX builds for decode speed and keep GGUF when the MLX catalog lacks a quant you need. Treat Bionic Secure Cloud as burst capacity for frontier open models, not the default; local MLX remains the privacy and cost baseline. The CEO has said publicly that the team builds Bionic with Bionic using models like Kimi K3, which is a signal the agent path is meant for real coding work, not a demo shell. Many power users still keep Ollama installed for scripts while using LM Studio to browse, compare, and serve to editors on port 1234. ## FAQ ### Is LM Studio free in 2026? Yes. The desktop app is free for personal and work use. You only pay if you optionally use LM Studio Secure Cloud models inside Bionic or if you buy hardware capable of running large models. ### Does LM Studio support Apple Silicon and macOS Tahoe? Yes. Apple Silicon is a first-class target via MLX and Metal-accelerated llama.cpp paths. Release notes through 2026 reference modern macOS support including macOS 26-era fixes; install the latest build for Tahoe. ### What is LM Studio Bionic? Bionic is LM Studio's agent app (launched July 2026) for coding and document work with open models: local, LM Link, or Secure Cloud, with voice transcription and Zero Data Retention on cloud calls. It is separate from the classic LM Studio chat/server app. ### Which models can I run? Any supported GGUF or MLX open model you download. Examples discussed in 2026 materials include Llama-family builds, Gemma, Qwen, DeepSeek variants, gpt-oss, GLM 5.2, Kimi K3, and many community quantisations. Availability depends on format support in your engine version. ### LM Studio vs Ollama on Mac? Choose LM Studio for GUI model management, MCP UI, and Bionic agent workflows. Choose Ollama for minimal CLI workflows and scripting. Many power users run both. ### Can I call local models from other apps? Yes. Enable the local server and point OpenAI-compatible clients, SDKs, or Claude Code's Anthropic-compatible path at LM Studio. ### Is my data private? Local chats and model weights stay on disk. Cloud features in Bionic are optional and advertised as Zero Data Retention; review current LM Studio privacy docs before enabling cloud models. ## Sources - [Introducing LM Studio 0.4.0](https://lmstudio.ai/blog/0.4.0) - [Introducing LM Studio Bionic](https://lmstudio.ai/blog/introducing-lm-studio-bionic) - [LM Studio home](https://lmstudio.ai/) - [LM Studio is free for use at work](https://lmstudio.ai/blog/free-for-work) - [Apple Silicon LLMs complete guide 2026](https://codersera.com/blog/apple-silicon-llms-complete-guide-2026/) - [LM Studio and 5 Alternatives for Local AI](https://elephas.app/blog/lm-studio-review) - [Multi-Model Routing with LM Studio and MLX](https://medium.com/@michael.hannecke/the-same-router-better-backend-multi-model-routing-with-lm-studio-and-apples-mlx-78f53b2aabbb) - [Locally AI joins LM Studio](https://lmstudio.ai/blog/locally-ai-joins-lm-studio) ## Related - [Cursor](https://bundl.run/apps/cursor) - [Claude Code](https://bundl.run/apps/claude-code) - [ChatGPT](https://bundl.run/apps/chatgpt) - [Claude](https://bundl.run/apps/claude) - [Codex](https://bundl.run/apps/codex) - [Windsurf](https://bundl.run/apps/windsurf) - [LM Studio vs Ollama](https://bundl.run/compare/ollama-vs-lm-studio) - [LM Studio vs Open WebUI](https://bundl.run/compare/lm-studio-vs-open-webui) - [Free alternative to Private LLM](https://bundl.run/https:/) ```json { "@context": "https://schema.org", "@graph": [ { "@type": "Organization", "@id": "https://bundl.run/#organization", "name": "Bundl.run", "url": "https://bundl.run", "logo": { "@type": "ImageObject", "url": "https://bundl.run/og-image.png", "width": 1200, "height": 630 }, "description": "The Ninite for Mac. Install all your essential Mac apps with one terminal command.", "sameAs": [ "https://github.com/abhiofficial/bundl-mac-setup", "https://x.com/bundlrun", "https://www.producthunt.com/products/bundl-run" ], "foundingDate": "2024", "contactPoint": { "@type": "ContactPoint", "contactType": "customer support", "url": "https://bundl.run/faq" } }, { "@type": "WebSite", "@id": "https://bundl.run/#website", "name": "Bundl.run", "url": "https://bundl.run", "description": "The Ninite for Mac. Install all your essential Mac apps with one terminal command.", "publisher": { "@id": "https://bundl.run/#organization" }, "inLanguage": "en-US" }, { "@type": "Person", "@id": "https://bundl.run/authors/alex-chen#person", "name": "Alex Chen", "jobTitle": "Senior Developer Tools Specialist", "url": "https://bundl.run/authors/alex-chen", "worksFor": { "@id": "https://bundl.run/#organization" }, "description": "Alex Chen has been evaluating developer tools and productivity software for over 12 years, with deep expertise in code editors, terminal emulators, and development environments. As a former software engineer at several Bay Area startups, Alex brings hands-on experience with the real-world workflows these tools are meant to enhance. Alex tests each application extensively on both Intel and Apple Silicon Macs, documenting performance metrics, integration capabilities, and workflow efficiency. When not reviewing software, Alex contributes to open-source projects and writes technical tutorials for the developer community.", "knowsAbout": [ "Code Editors & IDEs", "Terminal Emulators", "Version Control Tools", "DevOps & CI/CD", "API Development", "Performance Benchmarking" ], "image": "https://bundl.run/authors/alex-chen.svg" }, { "@type": "BreadcrumbList", "@id": "https://bundl.run/apps/lm-studio#breadcrumb", "itemListElement": [ { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://bundl.run" }, { "@type": "ListItem", "position": 2, "name": "Apps", "item": "https://bundl.run/apps" }, { "@type": "ListItem", "position": 3, "name": "LM Studio", "item": "https://bundl.run/apps/lm-studio" } ] }, { "@type": "WebPage", "@id": "https://bundl.run/apps/lm-studio", "url": "https://bundl.run/apps/lm-studio", "name": "LM Studio for Mac", "description": "Discover, download, and run local LLMs with a desktop GUI", "isPartOf": { "@id": "https://bundl.run/#website" }, "publisher": { "@id": "https://bundl.run/#organization" }, "inLanguage": "en-US", "datePublished": "2026-01-01T00:00:00Z", "dateModified": "2026-08-10T13:34:04.000Z", "author": { "@id": "https://bundl.run/authors/alex-chen#person" }, "mainEntity": { "@id": "https://bundl.run/apps/lm-studio#software" }, "breadcrumb": { "@id": "https://bundl.run/apps/lm-studio#breadcrumb" } }, { "@type": "SoftwareApplication", "@id": "https://bundl.run/apps/lm-studio#software", "name": "LM Studio", "description": "Discover, download, and run local LLMs with a desktop GUI", "applicationCategory": "DeveloperApplication", "operatingSystem": "macOS", "url": "https://bundl.run/apps/lm-studio", "image": { "@type": "ImageObject", "url": "https://img.logo.dev/lmstudio.ai", "caption": "LM Studio app icon for Mac" }, "sameAs": [ "https://lmstudio.ai/", "https://formulae.brew.sh/cask/lm-studio" ], "offers": { "@type": "Offer", "price": "0", "priceCurrency": "USD", "availability": "https://schema.org/InStock" }, "review": { "@type": "Review", "author": { "@id": "https://bundl.run/authors/alex-chen#person" }, "reviewBody": "In August 2026 LM Studio is the default GUI for serious local LLM work on Mac, now paired with Bionic for agentic coding and docs. Free-for-work licensing, MLX performance, and a rapid release train make it hard to beat if you want open models under your control." }, "about": { "@type": "Thing", "name": "LM Studio", "description": "Discover, download, and run local LLMs with a desktop GUI" }, "isPartOf": { "@id": "https://bundl.run/#website" } }, { "@type": "Article", "@id": "https://bundl.run/apps/lm-studio#article", "headline": "LM Studio for Mac — Full Review & Installation Guide 2026", "description": "In August 2026 LM Studio is the default GUI for serious local LLM work on Mac, now paired with Bionic for agentic coding and docs. Free-for-work licensing, MLX performance, and a rapid release train make it hard to beat if you want open models under your control.", "image": "https://img.logo.dev/lmstudio.ai", "author": { "@id": "https://bundl.run/authors/alex-chen#person" }, "publisher": { "@id": "https://bundl.run/#organization" }, "datePublished": "2026-01-01T00:00:00Z", "dateModified": "2026-08-10T13:34:04.000Z", "mainEntityOfPage": { "@type": "WebPage", "@id": "https://bundl.run/apps/lm-studio" }, "articleSection": "DeveloperApplication", "speakable": { "@type": "SpeakableSpecification", "cssSelector": [ "h1", ".key-facts" ] }, "about": { "@type": "SoftwareApplication", "name": "LM Studio", "url": "https://bundl.run/apps/lm-studio" }, "mentions": [ { "@type": "SoftwareApplication", "name": "Private LLM" } ] }, { "@type": "FAQPage", "@id": "https://bundl.run/apps/lm-studio#faq", "mainEntity": [ { "@type": "Question", "name": "Is LM Studio free in 2026?", "acceptedAnswer": { "@type": "Answer", "text": "Yes. The desktop app is free for personal and work use. You only pay if you optionally use LM Studio Secure Cloud models inside Bionic or if you buy hardware capable of running large models." } }, { "@type": "Question", "name": "Does LM Studio support Apple Silicon and macOS Tahoe?", "acceptedAnswer": { "@type": "Answer", "text": "Yes. Apple Silicon is a first-class target via MLX and Metal-accelerated llama.cpp paths. Release notes through 2026 reference modern macOS support including macOS 26-era fixes; install the latest build for Tahoe." } }, { "@type": "Question", "name": "What is LM Studio Bionic?", "acceptedAnswer": { "@type": "Answer", "text": "Bionic is LM Studio's agent app (launched July 2026) for coding and document work with open models: local, LM Link, or Secure Cloud, with voice transcription and Zero Data Retention on cloud calls. It is separate from the classic LM Studio chat/server app." } }, { "@type": "Question", "name": "Which models can I run?", "acceptedAnswer": { "@type": "Answer", "text": "Any supported GGUF or MLX open model you download. Examples discussed in 2026 materials include Llama-family builds, Gemma, Qwen, DeepSeek variants, gpt-oss, GLM 5.2, Kimi K3, and many community quantisations. Availability depends on format support in your engine version." } }, { "@type": "Question", "name": "LM Studio vs Ollama on Mac?", "acceptedAnswer": { "@type": "Answer", "text": "Choose LM Studio for GUI model management, MCP UI, and Bionic agent workflows. Choose Ollama for minimal CLI workflows and scripting. Many power users run both." } }, { "@type": "Question", "name": "Can I call local models from other apps?", "acceptedAnswer": { "@type": "Answer", "text": "Yes. Enable the local server and point OpenAI-compatible clients, SDKs, or Claude Code's Anthropic-compatible path at LM Studio." } }, { "@type": "Question", "name": "Is my data private?", "acceptedAnswer": { "@type": "Answer", "text": "Local chats and model weights stay on disk. Cloud features in Bionic are optional and advertised as Zero Data Retention; review current LM Studio privacy docs before enabling cloud models." } } ], "isPartOf": { "@id": "https://bundl.run/apps/lm-studio" } } ] } ```