The Open Source Alternative to Algolia + Pinecone

The Easier To Use Alternative to Elasticsearch

 
Search meets 2024.

Typesense is a modern, privacy-friendly, open source search engine meticulously engineered for performance & ease-of-use.

It uses cutting-edge search algorithms that take advantage of the latest advances in Hardware Capabilities & Machine Learning.

  • Search-as-you-type
  • Autocomplete
  • Faceted Navigation
  • Geo-Search
  • Fuzzy Search
  • Vector Search
  • Semantic Search
  • Recommendations
  • LLM Augmentation
  • Geo-Distributed Cache
Read more about Typesense's use-cases
 
 

Trusted by Teams of all Sizes

You'll be in good company
CodecademyLogitechChangelogElevenLabsOddle EatsMacro FactorPoorvika MobilesAeroTradeSoundsnapTrading Strategy
 

As Seen At

Typesense at Google/IO
Typesense on ThoughtWorks Technology Radar
Typesense on Nasdaq Billboard
Typesense on Nasdaq Billboard
 

Features you love, minus the complexity

Neatly wrapped in an intuitive RESTful API
Typo Tolerance

Spellng Mistakes? Not a problem. Typesense automatically tries to correct typos.

Tunable Ranking

Tailor your results to perfection via flexible and fast query-time ranking.

Merchandising

Pin specific records in a particular position to feature or merchandize them.

Synonyms

Show results for pants when users search for trousers, or vice-versa, when you define them as synonyms.

Multi-tenant API Keys

Store multiple users’ data in a single index, create API keys for each user that restrict access to just their data.

Dynamic Sorting

Sort records on the fly by any fields in your document. For eg: sort by price, sort by popularity, etc. No duplicate indices needed.

Grouping & Distinct

Provide more variety in your results by grouping results. Eg: combine all color variations of a shirt into a single result.

Filtering & Faceting

Only fetch records that match a given filter. Aggregate field values and get counts, min, max and avg of values across records.

Geo Search

Search & sort results within a certain distance from a latitude/longitude or within a polygon region.

Federated Search

Search one or more collections in a single query. Eg: search for both products and brands, given a single search query.

Vector & Semantic Search

Automatically generate embeddings using built-in ML models or OpenAI / PaLM API and do semantic search or nearest-neighbor search.

Easy High Availability

Build a resilient production-grade search service, with a few simple steps.

 

Batteries-Included Developer Experience

Zero to Instant-Search, in 30 seconds 🚀

1. Run Typesense

Use Docker, our native binaries or our RPM or DEB packages.
Or, use Typesense Cloud to provision a cluster with a few clicks.

2. Index Data

Create a collection, then add some documents to it.

And, that's it! Search away.

You now have a typo-tolerant, search server you can send search queries to.
 

API Libraries & Plugins in your Favorite Languages

And if you don't find a library, reach out to us to collab on one
Javascript Golang PHP Ruby Python Dart Dot Net Clojure Java Swift Rust
React Vue Angular Laravel Symfony
 
Gatsby InstantSearch.js DocSearch DocSearch
Firebase Extension Wordpress WooCommerce Magento Craft CMS Prepr CMS Plone CMS Tooljet Extension LangChain
 

magic wand
Compose Powerful Search & Discovery User Interfaces, quickly.

InstantSearch.js + Typesense UI Adapter = Awesomeness
 
 

Backed by responsive support

Let us help you build a scalable and robust search experience
 
 

magic wand
We also offer Concierge Onboarding and Prioritized Support, for when you need it.