AI Toolkit for Visual Studio Code is a powerful extension for Visual Studio Code that streamlines agent development. AI Engineers can easily build AI applications by developing and testing generative AI models—locally or in the cloud. The extension supports most major generative models available today.
With AI Toolkit, AI Engineers can:
-
Explore and evaluate models from a wide range of providers—including Anthropic, OpenAI, GitHub—or run models locally using ONNX and Ollama
-
Build and test agents in minutes with prompt generation, quick starters, and seamless MCP tool integrations.
-
Fine-tune models on a local machine with GPU or in the cloud (Azure Container App) with GPU
The AI Toolkit samples are designed to help developers and researchers explore and implement AI solutions effectively. Our samples include:
- Sample Code: Pre-built examples to demonstrate AI functionalities, such as training, deploying, or integrating models into applications.
- Documentation: Guides and tutorials to help users understand AI Toolkit features and how to use them.
We have a dedicatd AI Toolkit Channel in our Azure AI Foundry Discord Community. Join Now at
Prequisites
- Visual Studio Code
- AI Toolkit for Visual Studio Code
- GitHub Fine-grained personal access token (PAT)
Get Started
- Star and fork this repository.
- In Visual Studio Code, clone your forked repository. You can do so via the Command Palette (CTRL/CMD+Shift+P) by first searching for Git: Clone and selecting your fork to clone locally. (Note: Alternatively, you can run the command
git clone https://github.com/Azure-Samples/AI_Toolkit_Samples.gitin the terminal.) - After cloning the repo, open the project folder in VS Code and navigate to the Samples folder.
Want to dive deeper into what you can do with the AI Toolkit? Our blog articles cover everything from getting started to advanced use cases, including tips for working with local and cloud-based models, prompt engineering techniques, and evaluation strategies.
Whether you're just exploring or ready to build production-ready AI apps, these posts are designed to help you get the most out of the extension.
Check out the latest articles below to expand your skills and stay up to date with new features!
| Article | Abstract |
|---|---|
| Visual Studio AI Toolkit : Building Phi-3 GenAI Applications | Learn how to run pre-optimized AI models locally, test and integrate them seamlessly, and fine-tune models for specific needs. |
| Expanded model catalog for AI Toolkit | The AI Toolkit now supports a broader range of models, including those from Google, Anthropic, and GitHub, enhancing its capabilities for integrating AI into applications. |
| Bring your own models | The AI Toolkit now supports local models via Microsoft Foundry Local, Ollama and remote models using API keys for OpenAI, Google, and Anthropic, expanding its versatility for developers. |
| Prompt Engineering Simplified: AI Toolkit's Prompt Builder | The Prompt Builder in AI Toolkit simplifies the creation and refinement of prompts for large language models, enabling users to generate, test, and customize prompts efficiently. |
| Building Retrieval Augmented Generation on VSCode & AI Toolkit | Learn how to build a Retrieval-Augmented Generation (RAG) application using VS Code and AI Toolkit. |
| Building RAG on Phi-3 locally using embeddings on VS Code AI Toolkit | Learn how to build a Retrieval-Augmented Generation (RAG) application locally using the Phi-3 model and embeddings with the AI Toolkit. |
| How Reasoning Models are transforming Logical AI thinking | Explore how reasoning models excel at logical problem-solving by breaking down complex tasks into smaller steps and solving them through explicit logical reasoning. |
| Recipe Generator Application with Phi-3 Vision on AI Toolkit Locally | Learn how to create a Recipe Generator Application using the Phi-3 Vision model and the AI Toolkit. |
SLMs & Local Models – Test and deploy AI models and applications efficiently on your own terms locally, to edge devices or to the cloud Embedding Models & RAG – Supercharge retrieval for smarter applications using existing data. Multi-Modal AI – Work with images, text, and beyond. Agentic Frameworks – Build autonomous, decision-making AI systems.
What will you learn from this session? Whether you're a developer, startup founder, or AI enthusiast, you'll gain practical insights, live demos, and actionable takeaways to level up your AI integration journey.
Join us and spark your AI transformation - view the AI Sparks playlist!
Thinking about including the AI Toolkit in a presentation? We’ve got you covered with a set of ready-to-go demos to help you learn and showcase both the AI Toolkit and Azure AI Foundry extensions.
| Website | Page Title |
|---|---|
| July Update | AI Toolkit July 2025 Update |
| June Update | AI Toolkit June 2025 Update |
| March Update | AI Toolkit March 2025 Update |
| Feb Update | AI Tookit Feb 2025 Update |
| Jan Update | AI Toolkit Jan 2025 Update |
| Oct Update | AI Toolkit Oct 2024 Update |