Azure AI + Vercel AI SDK RAG Template

A Next.js application, powered by the Vercel AI SDK, that uses retrieval-augmented generation (RAG) to reason and respond with information outside of the model's training data.

Features

• Information retrieval and addition through tool calls using the streamText function
• Real-time streaming of model responses to the frontend using the useChat hook
• Vector embedding retrieval with Azure AI Search using embeddings from Azure OpenAI
• Generative text streaming with Azure OpenAI
• Animated UI with Framer Motion

Getting Started

To get the project up and running, follow these steps:

1. Install dependencies:

   npm install
   

2. Copy the example environment file:

   cp .env.example .env
   

3. Create prerequisite resources in Azure AI:

• Azure AI Search index
  • Optionally create a semantic search configuration and include vector fields in the index

Note: create a vector search index via REST API or within Azure Portal)

• Azure OpenAI Chat model
• Azure OpenAI Embedding model, if using vector search

Note: see available Azure OpenAI models here and deploy them using this guide

1. Add your Azure OpenAI and Azure AI Search variables to the .env file:

   AZURE_SEARCH_ENDPOINT=your_azure_search_endpoint_here
   AZURE_SEARCH_KEY=your_azure_search_key_here
   AZURE_SEARCH_INDEX_NAME=your_azure_search_index_name_here
   AZURE_SEARCH_CONTENT_FIELD=your_azure_search_content_field_here
   AZURE_SEARCH_VECTOR_FIELD=your_azure_search_vector_field_here # include if using vector search
   AZURE_SEARCH_SEMANTIC_CONFIGURATION_NAME=your_azure_search_semantic_configuration_name_here # include if using semantic search
   
   AZURE_OPENAI_API_ENDPOINT=your_azure_openai_api_endpoint_here
   AZURE_RESOURCE_NAME=your_azure_resource_name_here
   AZURE_DEPLOYMENT_NAME=your_azure_deployment_name_here # chat model deployment name
   AZURE_EMBEDDING_DEPLOYMENT_NAME=your_azure_embedding_deployment_name_here # embedding model deployment name
   AZURE_API_KEY=your_azure_api_key_here
   

2. Start the development server:

   npm run dev
   

Your project should now be running on http://localhost:3000.

Virtual Dev Environments

Appendix

This template includes code from Vercel's AI SDK RAG Template, which is licensed under the MIT license.

Trademarks

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft’s Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party’s policies.