Supavec

About Supavec

Supavec is an open-source RAG-as-a-service API: you upload text or documents, it embeds and stores them, and you query for the passages that answer a question, with the whole thing reachable over a REST call rather than a vector-database deployment of your own. It is built on Supabase row-level security, which is the design decision that carries most of the weight — tenant isolation is enforced by the database rather than by application code you have to get right, so a multi-tenant product does not leak one customer's documents into another customer's answers. The positioning is explicitly as the open-source alternative to Carbon.ai, and the self-host path is real rather than decorative. The use cases the team leads with are the ones where retrieval quality matters more than model choice: sales-call transcripts you can ask about pricing objections and get timestamped clips back, a support knowledge copilot fed from Zendesk, Notion or Confluence that answers with citations to the current doc, on-premise legal and HR policy Q&A queried through Slack with exact clause references, and hourly-synced documentation search that returns code snippets in under 300 ms. Getting started is a POST to an upload endpoint with a bearer token; SDKs exist for React, Python and Node. There is a free forever tier and a 14-day no-questions refund on paid plans.

Key Features

Tags

More LLM APIs & Models tools