azure-search-openai-javascript
A TypeScript sample app for the Retrieval Augmented Generation pattern running on Azure, using Azure AI Search for retrieval and Azure OpenAI and LangChain larg
What is azure-search-openai-javascript?
A TypeScript sample app for the Retrieval Augmented Generation pattern running on Azure, using Azure AI Search for retrieval and Azure OpenAI and LangChain large language models (LLMs) to power ChatGPT-style and Q&A experiences.
Overview of azure-search-openai-javascript
What it is
It is a TypeScript sample application that implements the Retrieval Augmented Generation (RAG) pattern on Azure. It uses Azure AI Search for data retrieval and Azure OpenAI large language models (gpt-4o-mini) to provide ChatGPT-style and Q&A experiences over enterprise data. The sample includes fictitious Contoso Real Estate documents covering terms of service, privacy policy, and a support guide.
Key features
- Chat and Q&A interfaces with citations and source tracking
- Data preparation, prompt construction, and orchestration between the model and retriever
- Settings in the UX to tweak behavior and experiment with options
- Optional performance tracing and monitoring with Application Insights
Use cases
- Building customer support chatbots that answer questions using internal documents
- Creating internal knowledge bases with retrieval-augmented question answering
- Prototyping and evaluating RAG patterns with Azure AI Search and Azure OpenAI
Getting started
See the repository README for detailed setup instructions. The repository describes Azure account prerequisites and multiple deployment methods including GitHub Codespaces, VS Code Remote Containers, and local environment setup.
License and pricing
License: MIT (open source). Costs depend on the Azure resources used; the README notes that exact costs cannot be estimated and recommends the Azure pricing calculator. The project is listed as open source with no additional pricing.
Based on the project's public README and repository metadata.