Overview for the AI Toolkit for Visual Studio Code
The AI Toolkit for Visual Studio Code (VS Code) is a VS Code extension that simplifies generative AI app development by bringing together cutting-edge AI development tools and models from the Azure AI Foundry catalog and other catalogs like Hugging Face.
Note
Additional documentation and tutorials for the AI Toolkit for VS Code are available in the VS Code documentation: AI Toolkit for Visual Studio Code. You’ll find guidance on Playground, working with AI models, fine-tuning local and cloud-based models, and more.
The AI Toolkit for VS Code (AI Toolkit) enables you to:
- Download and run AI models locally. The AI Toolkit provides access to highly optimized models for the following platforms and hardware:
- Windows 11 running with DirectML acceleration
- Windows 11 running directly on the CPU
- Windows 11 Copilot+ PCs running on the NPU
- Linux with NVIDIA acceleration
- Linux running directly on the CPU
- MacOS support is coming soon
- Test models in an intuitive playground or in your application with a REST API.
- Fine-tune your AI model – locally or in the cloud (on a virtual machine) – to create new skills, improve reliability of responses, set the tone and format of the response. The AI Toolkit provides a guided walkthrough to fine-tune popular small-language models (SLMs) – like Phi-3 and Mistral.
- Deploy your AI feature either to the cloud or with an application that runs on a device.
- Batch run prompts for selected AI models in the Playground.
Want to learn more about the AI Toolkit for VS Code? Dona Sarkar and John Lam provide an overview of the AI Toolkit in this video from Microsoft Build 2024:
For the latest documentation and to download and use the AI Toolkit, please visit the VS Code documentation.
