{"id":59832,"date":"2025-06-19T15:49:00","date_gmt":"2025-06-19T13:49:00","guid":{"rendered":"https:\/\/www.schneider.systems\/?p=59832"},"modified":"2025-06-19T16:04:24","modified_gmt":"2025-06-19T14:04:24","slug":"microsoft-foundry-local-run-ai-models-locally","status":"publish","type":"post","link":"https:\/\/www.schneider.systems\/lu\/microsoft-foundry-local-run-ai-models-locally\/","title":{"rendered":"Microsoft Foundry Local: Run AI models locally"},"content":{"rendered":"<p>On May 19, 2025, Microsoft <strong>Foundry Local<\/strong> was launched in <strong>public preview<\/strong>. It enables you to <strong>run AI models directly on Windows and macOS devices<\/strong>. This <strong>eliminates cloud dependencies<\/strong> while maximizing performance, privacy, and control over sensitive data.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-full wp-image-59837\" src=\"https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Website-Microsoft-Foundry-Local.jpg\" alt=\"\" width=\"800\" height=\"449\" srcset=\"https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Website-Microsoft-Foundry-Local.jpg 800w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Website-Microsoft-Foundry-Local-500x281.jpg 500w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Website-Microsoft-Foundry-Local-150x84.jpg 150w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Website-Microsoft-Foundry-Local-768x431.jpg 768w\" sizes=\"auto, (max-width: 800px) 100vw, 800px\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2><strong>What Is Foundry Local?<\/strong><\/h2>\n<p><strong>Foundry Local<\/strong> is Microsoft\u2019s new <strong>local AI platform<\/strong> that brings the <strong>advanced capabilities of <a href=\"https:\/\/www.schneider.systems\/lu\/microsoft-azure-ai-foundry-the-new-azure-ai-portal\/\">Azure AI Foundry<\/a> directly to your desktop or on-premises server<\/strong>. Whether you&#8217;re a developer or part of an enterprise team, Foundry Local helps you <strong>build fast, secure, and intelligent apps\u2014without needing to connect to the cloud.<\/strong><\/p>\n<h3><strong>Key Features:<\/strong><\/h3>\n<ul>\n<li><strong>Run AI Models Right on Your Device<\/strong>: Foundry Local lets you run models directly on your laptop or desktop, keeping everything on your own hardware.<\/li>\n<li><strong>Use Ready-Made Models or Bring Your Own<\/strong>: You can choose from prebuilt AI models or upload your own, depending on what your project needs.<\/li>\n<li><strong>Save Money by Using What You Already Have<\/strong>: Since Foundry Local runs on your existing devices, you avoid ongoing cloud service fees.<\/li>\n<li><strong>Easy to Connect with Your Apps<\/strong>: Foundry Local works smoothly with your tools using a command-line interface (CLI), software development kit (SDK), or REST API. And if your needs grow, you can scale up to Azure AI Foundry in the cloud.<\/li>\n<li><strong>Works on Windows and macOS<\/strong>: Whether you are on a PC or a Mac, you can build and run your AI apps across both platforms.<\/li>\n<li><strong>Especially great for Testing: <\/strong>You can Foundry Local for testing your AI models locally, and when you are ready, move them to the cloud for broader deployment.<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2><strong>How It Works (Architecture)<\/strong><\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-large wp-image-59833\" src=\"https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Microsoft-Foundry-Local-Architecture-800x397.png\" alt=\"\" width=\"800\" height=\"397\" srcset=\"https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Microsoft-Foundry-Local-Architecture-800x397.png 800w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Microsoft-Foundry-Local-Architecture-500x248.png 500w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Microsoft-Foundry-Local-Architecture-150x74.png 150w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Microsoft-Foundry-Local-Architecture-768x381.png 768w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Microsoft-Foundry-Local-Architecture-1536x762.png 1536w, https:\/\/www.schneider.systems\/media\/2025\/06\/SCHNEIDER-IT-MANAGEMENT-2025-06-19-Update-Microsoft-Foundry-Local-Architecture-2048x1016.png 2048w\" sizes=\"auto, (max-width: 800px) 100vw, 800px\" \/><\/p>\n<ul>\n<li><strong>Foundry Local Service<\/strong> runs on your device and provides an OpenAI-compatible API, allowing your apps to send requests, run models, and receive results\u2014all locally.<\/li>\n<li><strong>ONNX Runtime<\/strong> is the engine that executes AI models efficiently on your hardware, whether it is a CPU, GPU, or NPU, and supports a wide range of devices from vendors like Intel, AMD, <a href=\"https:\/\/www.schneider.systems\/lu\/microsoft-azure-ai-and-nvidia-ai-partnership\/\">NVIDIA<\/a>, and Apple.<\/li>\n<li><strong>Model Cache<\/strong> stores downloaded models on your device so they\u2019re always ready to use without needing to re-download them, which speeds up inference.<\/li>\n<li><strong>CLI and SDK Access<\/strong> let developers manage models and run inference using simple commands or integrate Foundry Local directly into their applications.<\/li>\n<li><strong>Hardware Abstraction<\/strong> ensures Foundry Local automatically selects the best available hardware for each task, delivering fast and optimized performance across platforms.<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2><strong>Availability: Public Preview<\/strong><\/h2>\n<p><strong>Public preview<\/strong> is an early-access phase where the product is <strong>available to the public<\/strong> but is not yet considered fully production-ready. For Foundry Local, this means:<\/p>\n<ul>\n<li>You can download and use it now on Windows or macOS.<\/li>\n<li>It is functionally usable, but some features may still be evolving or limited.<\/li>\n<li>Microsoft may change APIs, behaviors, or capabilities before the General Availability (GA) release.<\/li>\n<li>However, it is not yet intended for distributed or containerized production deployments.<\/li>\n<\/ul>\n<p>The <strong>general availability release is yet to be announced<\/strong>.<\/p>\n<p>&nbsp;<\/p>\n<h2><strong>Licensing &amp; Cost<\/strong><\/h2>\n<ul>\n<li><strong>Foundry Local itself is free<\/strong> to install and use.<\/li>\n<li><strong>However, model usage may incur costs<\/strong>, depending on the specific model you choose to run. <strong>Some models are open-source and free, others may require a license<\/strong>.<\/li>\n<li><strong>You do not need an <a href=\"https:\/\/www.schneider.systems\/lu\/microsoft-azure-new-policy-on-unused-subscriptions-blocking-and-deletion\/\">Azure subscription<\/a><\/strong> to use Foundry Local\u2014it runs entirely on your local machine.<\/li>\n<\/ul>\n<h3><strong>How to Check If an Agent Incurs Cost<\/strong><\/h3>\n<p>You can <strong>use the Foundry command-line interface (CLI) <\/strong>to inspect the agent or model:<\/p>\n<pre style=\"padding-left: 40px;\"><code>foundry agent info background-remover --license<\/code><\/pre>\n<p>This will show you the licensing terms and whether any usage restrictions or fees apply.<\/p>\n<p>&nbsp;<\/p>\n<h2><strong>Installation &amp; Quickstart<\/strong><\/h2>\n<p>To install Foundry Local:<\/p>\n<h3><strong>Windows:<\/strong><\/h3>\n<pre style=\"padding-left: 40px;\"><code>winget install Microsoft.FoundryLocal<\/code><\/pre>\n<h3><strong>macOS:<\/strong><\/h3>\n<pre style=\"padding-left: 40px;\"><code>brew tap microsoft\/foundrylocal\r\nbrew install foundrylocal<\/code><\/pre>\n<h3><strong>To run your first model:<\/strong><\/h3>\n<pre style=\"padding-left: 40px;\"><code>foundry model run phi-3.5-mini<\/code><\/pre>\n<p>&nbsp;<\/p>\n<h2><strong>How to Pay for Licensed Models or Agents in Foundry Local<\/strong><\/h2>\n<p>As of now, Foundry Local itself is free, but <strong>licensed models or agents may require a separate agreement or license key<\/strong>.<\/p>\n<h3><strong>1. License Acceptance via CLI<\/strong><\/h3>\n<p>When you install or run a model or agent, Foundry Local may prompt you to <strong>accept a license<\/strong>. This is often done through the <strong>CLI<\/strong>:<\/p>\n<pre style=\"padding-left: 40px;\"><code>foundry model info &lt;model-name&gt; --license<\/code><\/pre>\n<p>If the model requires payment, the license terms will specify how to proceed\u2014this might include a link to a Microsoft or third-party licensing portal.<\/p>\n<h3><strong>2. Microsoft Marketplace or Partner Licensing<\/strong><\/h3>\n<p>For commercial models, Microsoft may route licensing through:<\/p>\n<ul>\n<li><strong>Azure Marketplace<\/strong><\/li>\n<li><strong>Third-party vendors<\/strong><\/li>\n<li><strong>Direct Microsoft licensing agreements<\/strong><\/li>\n<\/ul>\n<p>In these cases, you may be asked to:<\/p>\n<ul>\n<li>Purchase a <strong>license key<\/strong><\/li>\n<li>Subscribe to a <strong>usage plan<\/strong><\/li>\n<li><strong>Agree to terms<\/strong> for commercial deployment<\/li>\n<\/ul>\n<h3><strong>3. No Azure Subscription Required<\/strong><\/h3>\n<p>Even if the model is from Azure AI Foundry, you <strong>don\u2019t need an Azure subscription<\/strong> to run it locally. But if the model is licensed, you will <strong>still need to pay the model owner<\/strong>, not necessarily Microsoft.<\/p>\n<p>&nbsp;<\/p>\n<h2><strong>More information<\/strong><\/h2>\n<ul>\n<li>Learn about the integration of <strong>Foundry Local in Windows AI Foundry<\/strong>: <a href=\"https:\/\/blogs.windows.com\/windowsdeveloper\/2025\/05\/19\/advancing-windows-for-ai-development-new-platform-capabilities-and-tools-introduced-at-build-2025\/\">https:\/\/blogs.windows.com\/windowsdeveloper\/2025\/05\/19\/advancing-windows-for-ai-development-new-platform-capabilities-and-tools-introduced-at-build-2025\/<\/a>.<\/li>\n<li>Learn about <strong>Foundry Local architecture<\/strong>: <a href=\"https:\/\/learn.microsoft.com\/en-us\/azure\/ai-foundry\/foundry-local\/concepts\/foundry-local-architecture\">https:\/\/learn.microsoft.com\/en-us\/azure\/ai-foundry\/foundry-local\/concepts\/foundry-local-architecture<\/a>.<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2><strong>Need Help with Licensing?<\/strong><\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-large wp-image-56125\" src=\"https:\/\/www.schneider.systems\/media\/2025\/03\/SCHNEIDER-IT-MANAGEMENT-2025-03-11-Graphics-Office-People-9300-800x533.jpg\" alt=\"\" width=\"800\" height=\"533\" srcset=\"https:\/\/www.schneider.systems\/media\/2025\/03\/SCHNEIDER-IT-MANAGEMENT-2025-03-11-Graphics-Office-People-9300-800x533.jpg 800w, https:\/\/www.schneider.systems\/media\/2025\/03\/SCHNEIDER-IT-MANAGEMENT-2025-03-11-Graphics-Office-People-9300-500x333.jpg 500w, https:\/\/www.schneider.systems\/media\/2025\/03\/SCHNEIDER-IT-MANAGEMENT-2025-03-11-Graphics-Office-People-9300-150x100.jpg 150w, https:\/\/www.schneider.systems\/media\/2025\/03\/SCHNEIDER-IT-MANAGEMENT-2025-03-11-Graphics-Office-People-9300-768x512.jpg 768w, https:\/\/www.schneider.systems\/media\/2025\/03\/SCHNEIDER-IT-MANAGEMENT-2025-03-11-Graphics-Office-People-9300-1536x1024.jpg 1536w, https:\/\/www.schneider.systems\/media\/2025\/03\/SCHNEIDER-IT-MANAGEMENT-2025-03-11-Graphics-Office-People-9300-2048x1365.jpg 2048w\" sizes=\"auto, (max-width: 800px) 100vw, 800px\" \/><\/p>\n<p>To ensure compliance and optimize your AI licensing strategy, <a href=\"https:\/\/www.schneider.systems\/contact\/\">contact our licensing experts at SCHNEIDER IT MANAGEMENT<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>On May 19, 2025, Microsoft Foundry Local was launched in public preview. It enables you to run AI models directly on Windows and macOS devices. This eliminates cloud dependencies while [&hellip;]<\/p>\n","protected":false},"author":11,"featured_media":33016,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[248],"tags":[],"class_list":["post-59832","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-microsoft-azure"],"_links":{"self":[{"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/posts\/59832","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/users\/11"}],"replies":[{"embeddable":true,"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/comments?post=59832"}],"version-history":[{"count":3,"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/posts\/59832\/revisions"}],"predecessor-version":[{"id":59843,"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/posts\/59832\/revisions\/59843"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/media\/33016"}],"wp:attachment":[{"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/media?parent=59832"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/categories?post=59832"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.schneider.systems\/lu\/wp-json\/wp\/v2\/tags?post=59832"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}