oramasearch/orama

🌌 A complete search engine and RAG pipeline in your browser, server or edge network with support for full-text, vector, and hybrid search in less than 2kb.

What it solves

Orama provides a lightweight, flexible search engine that can be integrated directly into applications. It eliminates the need for complex external search infrastructure by offering full-text, vector, and hybrid search capabilities that can run in the browser, on the server, or in edge environments.

How it works

Users define a database schema with specific data types (such as strings, numbers, or vectors) and insert documents into the instance. Orama then enables various search modes:

  • Full-Text Search: Uses BM25 and supports typo tolerance, stemming, and tokenization across 30 languages.
  • Vector Search: Compares text embeddings to find semantically similar results.
  • Hybrid Search: Combines both full-text and vector search for improved accuracy.
  • Answer Engine: Integrates with LLMs (via plugins like Secure Proxy) to create RAG-based chat experiences where the engine provides context from the database to generate answers.

Who it’s for

Developers who need to add search functionality to their web or Node.js applications without managing a separate search server, as well as those building AI-powered chat interfaces and RAG systems.

Highlights

  • Versatile Search: Supports full-text, vector, hybrid, and geosearch.
  • Client-Side Ready: Can be imported via CDN and run directly in the browser.
  • RAG Support: Built-in AnswerSession for creating ChatGPT-like experiences using internal data.
  • Extensible: Features a plugin system for embeddings, data persistence, and secure API proxying.
  • Multilingual: Stemming and tokenization support for 30 languages.

Related

  • Dispatch
    Building LLM-Powered Web Apps with Client-Side Technology

    Ollama explores the feasibility of building fully local Retrieval-Augmented Generation (RAG) web applications using a combination of browser-based tools and Ollama for local LLM execution.

  • Project
    vespaai-playground/NyRAG

    A tool for building RAG applications by crawling websites or processing documents and deploying them to Vespa for hybrid search with an integrated chat UI.

  • Project
    vespa-engine/sample-apps

    A collection of reference applications and tutorials for Vespa, demonstrating how to implement hybrid search, RAG, and multimodal vector retrieval at scale.

  • Project
    Bessouat40/RAGLight

    A lightweight and modular Python library for implementing Retrieval-Augmented Generation (RAG), allowing users to easily integrate various LLMs and vector stores to chat with their documents.

  • Dispatch
    Ollama Embedding Models Support

    Ollama now supports embedding models, enabling the creation of local retrieval augmented generation (RAG) applications by converting text into vector embeddings.