meilisearch/meilisearch

📚 What is Meilisearch?

Meilisearch is an open‑source, lightning‑fast search engine designed to be easy to integrate into apps, websites, or any workflow. It provides full‑text, semantic (vector) and hybrid search through a simple RESTful API, with features like typo‑tolerance, filtering, faceting, sorting, synonyms, geosearch, multi‑tenancy, and AI‑ready integrations.

🚀 How to get started

Option	What you get	Quick start command
Meilisearch Cloud (recommended)	Managed service – no servers, built‑in monitoring, analytics, team management. 14‑day free trial.	Sign up at the Cloud portal and create a project.
Self‑hosted Community Edition	Open‑source binary you run yourself (Docker, curl install, Homebrew, APT, etc.).	sh\n docker run -it --rm -p 7700:7700 -v $(pwd)/meili_data:/meili_data getmeili/meilisearch:latest\n
Self‑hosted Enterprise Edition	Adds sharding and other advanced features; requires a commercial agreement.	See the Editions & licensing section.

Minimal example (Docker)

docker run -it --rm \
  -p 7700:7700 \
  -v $(pwd)/meili_data:/meili_data \
  getmeili/meilisearch:latest

Add documents via the API:

curl -X POST "http://localhost:7700/indexes/movies/documents" \
  -H 'Content-Type: application/json' \
  -d '[{"id":1,"title":"Carol"},{"id":2,"title":"Wonder Woman"}]'

Search (typo‑tolerant):

curl -X POST "http://localhost:7700/indexes/movies/search" \
  -H 'Content-Type: application/json' \
  -d '{"q":"wondre woman"}'

The response will include Wonder Woman despite the misspelling.

✨ Core features (highlighted)

• Hybrid search – combine vector (semantic) and classic full‑text search.
• Search‑as‑you‑type – results in < 50 ms.
• Typo tolerance – robust matching even with misspellings.
• Filtering & faceted search – build rich UI filters.
• Sorting, synonyms, geosearch – fine‑tune relevance.
• Multi‑tenancy & API‑key security – isolate tenants and control access.
• Replication & sharding – horizontal scaling.
• AI‑ready – native MCP endpoint, LangChain vector‑store support, conversational search, personalization, search rules, document relations.

🤖 AI & LLM integrations

• MCP server (/mcp route) – experimental endpoint for AI agents to query indexes directly.
• LangChain – use Meilisearch as a vector store in LangChain pipelines.
• Conversational search – natural‑language Q&A grounded in indexed data.
• llms.txt – machine‑readable index of the documentation for LLMs.

📦 SDKs & language support

Official SDKs exist for JavaScript, Python, PHP, Ruby, Go, Rust, React and many more. They wrap the REST API, making indexing and searching idiomatic in each language.

📚 Documentation & learning resources

• Full docs: https://www.meilisearch.com/docs
• API reference, guides on filtering, sorting, geosearch, security, multi‑tenancy, etc.
• Interactive demos (media search, agentic search, Flickr, e‑commerce, home‑booking, personalization, SaaS CRM).
• FAQ covering Cloud vs self‑host, EU hosting, comparisons to Algolia/Elasticsearch/Typesense.

🛠 Typical use‑cases

• E‑commerce product search – fast typo‑tolerant search with facets and filters.
• Media libraries – hybrid semantic + keyword search for movies, music, images.
• Enterprise SaaS – multi‑tenant search across customers.
• AI‑augmented apps – retrieve relevant documents for LLM prompts via MCP or LangChain.

🎯 Who should use it?

Anyone needing a quick‑to‑deploy, high‑performance search backend that can later be extended with AI‑driven features. It works for small startups (via Cloud) as well as larger teams that self‑host and scale with sharding.

---

All information above is taken directly from the repository’s README.