meilisearch/meilisearch

A lightning-fast search engine API bringing AI-powered hybrid search to your sites and applications.

What it solves

Meilisearch is a lightning-fast search engine designed to be integrated into apps and websites to provide a high-quality search experience. It addresses the challenge of implementing complex search functionality—such as typo tolerance, faceted search, and semantic search—without needing to build these systems from scratch.

How it works

Meilisearch operates as a standalone search engine accessible via a RESTful API. It allows developers to index documents and perform queries. It supports various search modalities, including full-text search, semantic search (vector search), and hybrid search (combining both), enabling results to be returned in under 50 milliseconds.

Who it’s for

It is intended for developers building applications that require fast, relevant, and customizable search interfaces, ranging from e-commerce sites to SaaS CRM applications and media libraries.

Highlights

  • Hybrid Search: Combines semantic and full-text search for maximum relevance.
  • Search-as-you-type: Delivers results in less than 50ms for an intuitive user experience.
  • Typo Tolerance: Ensures relevant matches even with misspellings.
  • Conversational Search: Enables natural language questions with AI-generated answers grounded in search results.
  • AI-Ready: Integrates out-of-the-box with LangChain and the Model Context Protocol (MCP).
  • Advanced Filtering: Supports faceted search, geosearch, and custom sorting.
  • Multi-Tenancy: Allows personalization of search results for different application tenants.

Related

  • Project
    meilisearch/meilisearch-dotnet

    A .NET API client for the Meilisearch open-source search engine, enabling C# developers to easily manage indexes and perform typo-tolerant searches.

  • Project
    RediSearch/RediSearch

    A Redis module that adds full-text search, secondary indexing, and vector similarity search to Redis, enabling high-performance querying of database records.

  • Project
    infiniflow/infinity

    An AI-native database designed for LLM applications that provides high-speed hybrid search across dense and sparse embeddings, tensors, and full-text data.

  • Project
    yokingma/SearChat

    An AI-powered conversational search engine that integrates multiple LLMs and search engines to provide real-time web search and iterative deep research reports.

  • Project
    typesense/typesense

    Typesense is an open-source, fast, typo-tolerant search engine built in C++ that serves as a simpler alternative to Elasticsearch and a self-hostable alternative to Algolia. It supports full-text, vector, semantic/hybrid, and conversational (RAG) search, plus image and voice search, all through a simple HTTP API.