Local runtime - Microsoft - verified
Microsoft Foundry Local: install, API, platforms and features
Microsoft Foundry Local is a local LLM runtime from Microsoft. As its docs describe it: Developers who want to ship on-device AI inside their own applications using native SDKs, a curated catalog of optimized ONNX models and automatic NPU, GPU or CPU acceleration for a single user per device. The latest release is v2.1.0, published September 30, 2026. License: Microsoft Software License Terms for the CLI, MIT for the SDK (not open source).
Homepage - Source on GitHub - Microsoft Software License Terms for the CLI, MIT for the SDK
- Latest release
- v2.1.0
- Released
- 2026-09-30
- License
- Not open source
- OpenAI API
- Yes
Install
Copied verbatim from the page linked under each command. Labels are worded as the source words them.
windows
winget install Microsoft.FoundryLocalSource: learn.microsoft.com/en-us/azure/foundry-local/reference/reference-cli
macos
brew tap microsoft/foundrylocal
brew install foundrylocalSource: learn.microsoft.com/en-us/azure/foundry-local/reference/reference-cli
javascript
npm install foundry-local-sdkRun a model
The first-run example from the official docs. OpenAI-compatible endpoint: Optional OpenAI-compatible local server started with `foundry server start` (or `start_web_service()` in the SDK); no fixed default port is documented, `--port <port>` sets one and `foundry server status` confirms it
foundry run qwen2.5-0.5bFeatures
What the Microsoft Foundry Local docs say it supports. Not stated means the docs do not say either way.
| Feature | Value | Source |
|---|---|---|
| Interface | CLI, Server, Python library | See sources below |
| Platforms | Windows, macOS, Linux | learn.microsoft.com/en-us/azure/foundry-local/what-is-foundry-local |
| Backends | CPU, CUDA, NPU, Other | learn.microsoft.com/en-us/azure/foundry-local/concepts/foundry-local-architecture |
| Model formats | ONNX | learn.microsoft.com/en-us/azure/foundry-local/concepts/foundry-local-architecture |
| OpenAI-compatible API | YesOptional OpenAI-compatible local server started with `foundry server start` (or `start_web_service()` in the SDK); no fixed default port is documented, `--port <port>` sets one and `foundry server status` confirms it | learn.microsoft.com/en-us/azure/foundry-local/what-is-foundry-local |
| Anthropic-compatible API | Not stated | Docs do not say |
| Tool calling | Yes | github.com/microsoft/Foundry-Local#-samples |
| Structured outputs | Not stated | Docs do not say |
| Vision (image input) | Not stated | Docs do not say |
| Multi-GPU | Not stated | Docs do not say |
| MCP client | Not stated | Docs do not say |
| License | Microsoft Software License Terms for the CLI, MIT for the SDKNot open source | github.com/microsoft/Foundry-Local/blob/main/LICENSE |
| Latest release | v2.1.0September 30, 2026 | github.com/microsoft/Foundry-Local/releases/tag/v2.1.0 |
Models to run with Microsoft Foundry Local
Browse the open-weight models you can run locally for download sizes, licenses and memory fit; check which formats Microsoft Foundry Local loads in the table above.
Compare Microsoft Foundry Local
FAQ
Does Microsoft Foundry Local have an OpenAI-compatible API?
Yes. Endpoint, per its docs: Optional OpenAI-compatible local server started with `foundry server start` (or `start_web_service()` in the SDK); no fixed default port is documented, `--port <port>` sets one and `foundry server status` confirms it. Source: https://learn.microsoft.com/en-us/azure/foundry-local/what-is-foundry-local
Does Microsoft Foundry Local have an Anthropic-compatible API?
Its official docs, read on October 18, 2026, do not say.
Which platforms and GPUs does Microsoft Foundry Local support?
Its docs list Windows, macOS and Linux, with these backends: CPU, CUDA, NPU and Other.
How do I install Microsoft Foundry Local?
On windows: "winget install Microsoft.FoundryLocal", as shown at https://learn.microsoft.com/en-us/azure/foundry-local/reference/reference-cli. This page lists 4 install commands: windows, macos, python and javascript.
Is Microsoft Foundry Local open source?
No. It is released under Microsoft Software License Terms for the CLI, MIT for the SDK; the license is at https://github.com/microsoft/Foundry-Local/blob/main/LICENSE.
What is the latest version of Microsoft Foundry Local?
v2.1.0, published September 30, 2026 (https://github.com/microsoft/Foundry-Local/releases/tag/v2.1.0), as of October 18, 2026.
Which model formats does Microsoft Foundry Local run?
ONNX, per https://learn.microsoft.com/en-us/azure/foundry-local/concepts/foundry-local-architecture
Sources
Last verified from the pages below.
- GitHub README - github.com
- Releases (v2.1.0 marked Latest) - github.com
- What is Foundry Local (Microsoft Learn) - learn.microsoft.com
- Architecture overview (Microsoft Learn) - learn.microsoft.com
- CLI reference (Microsoft Learn) - learn.microsoft.com
- Get started (Microsoft Learn) - learn.microsoft.com
Newsletter
Local AI, every week
One email per week. New model releases, runtime updates and what they mean for what you build.
Keep exploring
Open-weight models you can run
Download sizes, licenses and the run command for each model.
All local LLM runtimes
Every runtime in one feature matrix, with sources.
Picking a runtime for coding agents
Ollama, LM Studio, vLLM and llama.cpp for agent workloads.
Setup guides
Point your coding tools at the model you picked.