Azure Functions now
supports creating an OpenAI resource and an optional Azure AI Search vector
store for building intelligent applications in Flex consumption portal
experience. This makes it easier to build intelligent applications for text
completion, chat with assistant calling, or Retrieval augmented generation when
you want to use your own company data with Azure Open AI.
You can build intelligent
apps using the native Azure OpenAI SDKs or use the updated Azure Functions
OpenAI bindings to jumpstart development of the following scenarios:
Chat
assistants
- Input binding to chat with LLMs.
- Output binding to retrieve chat history from persisted
storage.
- Skills trigger to extend capabilities of OpenAI LLM through
natural language.
Retrieval
Augmented Generation (Bring your own data for semantic search)
- Data ingestion with Functions bindings.
- Automatic chunking and embeddings creation.
- Store embeddings in vector database including AI Search,
Cosmos DB for MongoDB, and Azure Data Explorer.
- Binding that takes prompts, retrieves documents, sends to
OpenAI LLM, and returns to user.
Text
completion for content summarization and creation
- Input binding that takes prompt or content and returns
response from LLM.
These bindings have
been improved in the latest preview to support managed identity and use the
latest underlying SDKs along with multiple fixes based on customer feedback.
Learn more.