DeepSpeed, ONNX, Phi, Semantic Kernel, Guidance
AI Agent Governance Toolkit — Policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering for autonomous AI agents. Covers 10/10 OWASP Agentic Top 10.
A programming framework for agentic AI
A modular graph-based Retrieval-Augmented Generation (RAG) system
Qlib is an AI-oriented Quant investment platform that aims to use AI tech to empower Quant Research, from exploring ideas to implementing productions. Qlib supports diverse ML modeling paradigms, including supervised learning, market dynamics modeling, and RL, and is now equipped with https://github.com/microsoft/RD-Agent to automate R&D process.
Open source simulator for autonomous vehicles built on Unreal Engine / Unity, from Microsoft AI & Research
Official inference framework for 1-bit LLMs
JARVIS, a system to connect LLMs with ML community. Paper: https://arxiv.org/pdf/2303.17580.pdf
A large-scale simulation framework for LLM inference
End-to-end Generative Optimization for AI Agents
Secure AI conversations with documents, video, audio, and more. Personal workspaces for focused context, group spaces for shared insight. Classify docs, reuse prompts, and extend with modular features.
A framework for building, orchestrating and deploying AI agents and multi-agent workflows with support for Python and .NET.
Curated lessons on getting started building agents with Copilot Studio
Windows Agent Arena (WAA) 🪟 is a scalable OS platform for testing and benchmarking of multi-modal AI agents.
This course is designed to guide beginners through the exciting world of Edge AI, covering fundamental concepts, popular models, inference techniques, device-specific applications, model optimization, and the development of intelligent Edge AI agents.
The Microsoft 365 Agent SDK simplifies building full stack, multichannel, trusted agents for platforms including M365, Teams, Copilot Studio, and Webchat.
12 Lessons to Get Started Building AI Agents
Skills, MCP servers, Custom Agents, Agents.md for SDKs to ground Coding Agents
MCP Gateway is a reverse proxy and management layer for MCP servers, enabling scalable, session-aware stateful routing and lifecycle management of MCP servers in Kubernetes environments.
Five lessons, learn how to really apply AI to your .NET Applications
Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are committed to automating these high-value generic R&D processes through R&D-Agent, which lets AI drive data-driven AI. 🔗https://aka.ms/RD-Agent-Tech-Report
Sample code that explores an architecture for using language models to build a personal agent that can work with application agents.
The Power BI Modeling MCP Server, brings Power BI semantic modeling capabilities to your AI agents.
General technology for enabling AI capabilities w/ LLMs and MLLMs
Hummingbird compiles trained ML models into tensor computation for faster inference.
A fast library for AutoML and tuning. Join our Discord: https://discord.gg/Cppx2vSPVP.
21 Lessons, Get Started Building with Generative AI
12 Weeks, 24 Lessons, AI for All!
This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".
The project is an official implement of our ECCV2018 paper "Simple Baselines for Human Pose Estimation and Tracking(https://arxiv.org/abs/1804.06208)"
[CVPR 2025] Magma: A Foundation Model for Multimodal AI Agents
The absolute trainer to light up AI agents.
Debugging, monitoring and visualization for Python Machine Learning and Data Science
TextWorld is a sandbox learning environment for the training and evaluation of reinforcement learning (RL) agents on text-based games.
Table Transformer (TATR) is a deep learning model for extracting tables from unstructured documents (PDFs and images). This is also the official repository for the PubTables-1M dataset and GriTS evaluation metric.
BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.
The Python Risk Identification Tool for generative AI (PyRIT) is an open source framework built to empower security professionals and engineers to proactively identify risks in generative AI systems.
[EMNLP'23, ACL'24] To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.
This is a Phi Family of SLMs book for getting started with Phi Models. Phi a family of open sourced AI models developed by Microsoft. Phi models are the most capable and cost-effective small language models (SLMs) available, outperforming models of the same size and next size up across a variety of language, reasoning, coding, and math benchmarks
Automatic Generation of Visualizations and Infographics using Large Language Models
UFO³: Weaving the Digital Agent Galaxy
LLM-powered multiagent persona simulation for imagination enhancement and business insights.
Integrate cutting-edge LLM technology quickly and easily into your apps
Official implementation of MatterGen -- a generative model for inorganic materials design across the periodic table that can be fine-tuned to steer the generation towards a wide range of property constraints.
Official Code Release for [SIGGRAPH 2025] RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination
A library of workshops written by and for Microsoft Learn Student Ambassadors and Cloud Advocates and their local communities
Graphormer is a general-purpose deep learning backbone for molecular modeling.
Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities
Azure MLOps (v2) solution accelerators. Enterprise ready templates to deploy your machine learning models on the Azure Platform.
MatterSim: A deep learning atomistic model across elements, temperatures and pressures.
PyTorch Wildlife: a Collaborative Deep Learning Framework for Conservation.
[ICLR 2025] Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling
Official community-driven Azure Machine Learning examples, tested with GitHub Actions.
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
PIKE-RAG: sPecIalized KnowledgE and Rationale Augmented Generation
Simple and Distributed Machine Learning
Automatable GenAI Scripting
12 weeks, 26 lessons, 52 quizzes, classic Machine Learning for all
FarmVibes.AI: Multi-Modal GeoSpatial ML Models for Agriculture and Sustainability
workshop materials to build intelligent solutions on Open AI
Low-bit LLM inference on CPU/NPU with lookup table
Join a time-traveling adventure where you meet history’s legends while learning Generative AI technologies! ✨
A library for helping developers craft prompts for Large Language Models
An open-source framework for detecting, redacting, masking, and anonymizing sensitive data (PII) across text, images, and structured data. Supports NLP, pattern matching, and customizable pipelines.
Learn Generative AI fundamentals through Java programming.
The first "code-first" agent framework for seamlessly planning and executing data analytics tasks.
A research prototype of a human-centered web agent
[arXiv 2023] Set-of-Mark Prompting for GPT-4V and LMMs
Sample code and documentation for using the Microsoft HoloLens 2 for Computer Vision research.
This solution accelerator leverages Microsoft Foundry, Azure Content Understanding, Azure OpenAI Service, and Foundry IQ to enable organizations to derive insights from volumes of conversational data using generative AI. It offers key phrase extraction, topic modeling, and interactive chat experiences through an intuitive web interface.
Inference code for scalable emulation of protein equilibrium ensembles with generative deep learning
AI-in-a-Box leverages the expertise of Microsoft across the globe to develop and provide AI and ML solutions to the technical community. Our intent is to present a curated collection of solution accelerators that can help engineers establish their AI/ML environments and solutions rapidly and with minimal friction.
End-to-End recipes for pre-training and fine-tuning BERT using Azure Machine Learning Service
The Multi-Agent Custom Automation Engine Solution Accelerator is an AI-driven system that manages a group of AI agents to accomplish tasks based on user input. Powered by Microsoft Agent Framework, Azure Foundry, Azure Cosmos DB, and infrastructure services, it provides a reference application, allowing you to hit the ground running.
Windows SDK for the Microsoft Computer Vision API, part of Cognitive Services
Microsoft Distributed Machine Learning Toolkit
a family of versatile and state-of-the-art video tokenizers.
Build your first agent with Azure AI Agent Service Workshop
MCP Python Tutorial
A highly-customizable web-based client for Azure Bot Services.
Android SDK for the Microsoft Computer Vision API, part of Cognitive Services.
A developer CLI for working with Azure resources to build and deploy AI applications. Commands map to key workflow stages: code, build, deploy, and monitor.
MarS: a Financial Market Simulation Engine Powered by Generative Foundation Model
Samples for getting started with deep learning across TensorFlow, CNTK, Theano and more.
Generation of protein sequences and evolutionary alignments via discrete diffusion models
Hack Together: RAG Hack | Register, Learn, Hack
Natural Language Processing Best Practices & Examples
O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis
Multi-Agent Resource Optimization (MARO) platform is an instance of Reinforcement Learning as a Service (RaaS) for real-world resource optimization problems.
Sample Code for Gated Graph Neural Networks
Unified-Modal Speech-Text Pre-Training for Spoken Language Processing
Prompt Orchestration Markup Language
A pre-trained GPT model for Python code completion and generation
FastFormers - highly efficient transformer models for NLU
A flexible and efficient deep neural network (DNN) compiler that generates high-performance executable from a DNN model description.
The IQ Series is a hands-on learning experience for Microsoft IQ: Microsoft's unified intelligence layer for the enterprise, spanning Foundry IQ, Work IQ, and Fabric IQ. The series includes video episodes, Jupyter notebooks, and Azure deployment templates.
Cookiecutter API for creating Custom Skills for Azure Search using Python and Docker
Gift your VS Code agent a real debugger: breakpoints, stepping, inspection.
This is an official implementation of CvT: Introducing Convolutions to Vision Transformers.
[NeurIPS 2021 Spotlight] Official code for "Focal Self-attention for Local-Global Interactions in Vision Transformers"
⚠️DirectML is in maintenance mode ⚠️ DirectML is a high-performance, hardware-accelerated DirectX 12 library for machine learning. DirectML provides GPU acceleration for common machine learning tasks across a broad range of supported hardware and drivers, including all DirectX 12-capable GPUs from vendors such as AMD, Intel, NVIDIA, and Qualcomm.
Research project. A Memory solution for users, teams, and applications.
Large-scale pretrained models for goal-directed dialog
A Dataset of Python Challenges for AI Research
Official Implementation of "KBLaM: Knowledge Base augmented Language Model"
Python notebooks with ML and deep learning examples with Azure Machine Learning Python SDK | Microsoft
CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped, CVPR 2022
Implementation of the Aurora model for Earth system forecasting
Microsoft Finance Time Series Forecasting Framework (FinnTS) is a forecasting package that utilizes cutting-edge time series forecasting and parallelization on the cloud to produce accurate forecasts for financial data.
Microsoft Cognitive Toolkit (CNTK), an open source deep-learning toolkit
multilspy is a lsp client library in Python intended to be used to build applications around language servers.
Build high-quality LLM apps - from prototyping, testing to production deployment and monitoring.
Prompty makes it easy to create, manage, debug, and evaluate LLM prompts for your AI applications. Prompty is an asset class and format for LLM prompts designed to enhance observability, understandability, and portability for developers.
A Unified Semi-Supervised Learning Codebase (NeurIPS'22)
This is a collection of our NAS and Vision Transformer work.
GIT: A Generative Image-to-text Transformer for Vision and Language
Sharing Updatable Models (SUM) on Blockchain
The RAG Experiment Accelerator is a versatile tool designed to expedite and facilitate the process of conducting experiments and evaluations using Azure Cognitive Search and RAG pattern.
Microsoft.Recognizers.Text provides recognition and resolution of numbers, units, date/time, etc. in multiple languages (ZH, EN, FR, ES, PT, DE, IT, TR, HI, NL. Partial support for JA, KO, AR, SV). Packages available at: https://www.nuget.org/profiles/Recognizers.Text, https://www.npmjs.com/~recognizers.text
Generative AI extensions for onnxruntime
Send a phone call from AI agent, in an API call. Or, directly call the bot from the configured phone number!
Agent Framework Samples - showcasing ways in which agent framework can be utilized.
TensorFlow implementations of Graph Neural Networks
TypeChat is a library that makes it easy to build natural language interfaces using types.
An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.
Sharing the learning along the way we been gathering to enable Azure OpenAI at enterprise scale in a secure manner. GPT-RAG core is a Retrieval-Augmented Generation pattern running in Azure, using Azure Cognitive Search for retrieval and Azure OpenAI large language models to power ChatGPT-style and Q&A experiences.
Official code for TimeCraft: A Time Series Generation Framework for Real-World Applications
Examples for using ONNX Runtime for machine learning inferencing.
Best Practices, code samples, and documentation for Computer Vision.
NLP DNN Toolkit - Building Your NLP DNN Models Like Playing Lego
Sample code and documentation for using the Microsoft HoloLens for Computer Vision research
Platform for Situated Intelligence
Diffusion models of protein structure; trigonometry and attention are all you need!
This repository provides code for machine learning algorithms for edge devices developed at Microsoft Research India.
A tutorial demonstrating how to implement deep learning models for time series forecasting
Spectral Temporal Graph Neural Network (StemGNN in short) for Multivariate Time-series Forecasting
MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.
RAG Time: A 5-week Learning Journey to Mastering RAG
A unified 3D Transformer Pipeline for visual synthesis
Resource scheduling and cluster management for AI
Automated benchmarking of Retrieval-Augmented Generation (RAG) systems
Samples and Tools for Windows ML.
Research code for CVPR 2021 paper "End-to-End Human Pose and Mesh Reconstruction with Transformers"
Generating native code interfaces from TypeScript
Multi-modality pre-training
Community for applying LLMs to robotics and a robot simulator with ChatGPT integration
[CVPR 2022] StyleSwin: Transformer-based GAN for High-resolution Image Generation
Large Language-and-Vision Assistant for Biomedicine, built towards multimodal GPT-4 level capabilities.
Microsoft Security Copilot is a generative AI-powered security solution that helps increase the efficiency and capabilities of defenders to improve security outcomes at machine speed and scale, while remaining compliant to responsible AI principles
Foundation Architecture for (M)LLMs
AICI: Prompts as (Wasm) Programs
Repo for Rho-1: Token-level Data Selection & Selective Pretraining of LLMs.
maximal update parametrization (µP)
For releasing code related to compression methods for transformers, accompanying our publications
This repo accelerates development of RAG applications with rich data sources including SQL Warehouses and documents analysed with Azure Document Intelligence
Experience, Learn and Code the latest breakthrough innovations with Microsoft AI
Parameter server framework for distributed machine learning
Medical Imaging Deep Learning library to train and deploy 3D segmentation models on Azure Machine Learning
Jupyter Notebook with Python samples for the Cognitive Services Computer Vision API
Visual Object Tagging Tool: An electron app for building end to end Object Detection Models from Images and Videos.
Muzic: Music Understanding and Generation with Artificial Intelligence
Tutel MoE: Optimized Mixture-of-Experts Library, Support GptOss/DeepSeek/Kimi-K2/Qwen3 using FP8/NVFP4/MXFP4