Back to directory

Nosia

A self-hosted Ruby on Rails platform for running AI models on your own data with an OpenAI-compatible API, retrieval-augmented generation over your documents, and Model Context Protocol integration to connect the assistant to external tools.

Nosia

Nosia lets an organization run AI models against its own private documents and data with a drop-in, OpenAI-compatible API that works with existing AI applications, keeping data on the organization's own infrastructure. Beyond standard retrieval-augmented generation over PDFs, text files, websites, and question-and-answer pairs, backed by pgvector semantic search, it integrates the Model Context Protocol so the assistant can call out to external tools and services, and supports custom Agent Skills written as LLM-driven prompts or plain Ruby code.

It supports account-based multi-tenancy for isolating data between users, streams responses over server-sent events, and deploys with a single Docker Compose command; installation scripts are provided for macOS, Linux, and Windows.

It is written in Ruby on Rails and released under the MIT license, with documentation covering architecture, system diagrams, deployment strategies, and custom skill development.

More in RAG

See all 540
  • ai·rete·rag

    ai-rete-rag.com

    A platform that pairs a Rete rule engine with retrieval-augmented generation, so deterministic rules decide an outcome and grounded retrieval explains the decision in plain language.

  • Xtriever

    github.com

    A hybrid retrieval engine for retrieval-augmented generation, written in Rust to run on-device on a phone or a laptop, combining lexical BM25 search, dense embedding search, reciprocal-rank fusion and cross-encoder re-ranking with no server.

  • FileFlux

    github.com

    A .NET library that turns PDF, DOCX, HWP and other documents into optimized chunks for RAG systems, through a five-stage pipeline built on high-performance Rust FFI readers.