EdgeDB 5: Introducing ext::ai

EdgeDB 5.0 introduces enhanced capabilities for indexing and searching content using semantic similarity through the ext::ai extension, building on the functionality of the ext::pgvector extension in version 3.0. The new version allows for seamless integration of semantic searches by merely declaring an index on a text property, facilitated by the deferred index mechanism that supports asynchronous index creation to minimize operation delays. This extension also enables integration with APIs from OpenAI, Mistral, and Anthropic, and can accommodate other models if they are API-compatible. The ext::ai extension supports retrieval-augmented generation (RAG) using database data, allowing developers to build AI-enabled applications efficiently. The @edgedb/ai JavaScript package simplifies interaction with the ext::ai HTTP API, promoting functionality like streaming outputs for enhanced user experience. Future developments aim to expand model support and improve integration with platforms like Vercel's AI SDK, complemented by guides and example projects. The updated EdgeDB UI now includes features for database interaction via conversation, facilitating a more intuitive user interface.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.