Search code, repositories, users, issues, pull requests...
Our service allows you to store and process both structured and unstructured data in a unified database format. Whether you're handling JSON objects, PDFs, or other document types, our npm package ensures all data is transformed and stored both as JSON entities and their vector embedding representations, always in-sync.
The interface for processing documents is similar to standard CRUD operations, making it easy to integrate. Regardless of whether you're storing traditional entities or processing documents, the outcome is the same: data is seamlessly written to the Farspeak database. This integration allows you to query all your data using natural language immediately, with structured and unstructured data coexisting harmoniously in a single database.
Farspeak supports both ESM and CommonJS modules, allowing you to integrate it seamlessly into your projects using either module system and you can use it with npm, yarn, bun, etc.
