DEV Community

Alex Spinov
Alex Spinov

Posted on

Typesense Has a Free Search Engine That Is Faster and Simpler Than Elasticsearch

#typesense #search #elasticsearch #database

Elasticsearch needs a JVM, 2GB+ RAM, and a PhD to configure. Typesense runs as a single binary, uses 100MB RAM, and returns results in under 50ms.

Why Typesense

Elasticsearch Meilisearch Typesense
Written in Java Rust C++
Min RAM 2GB 100MB 100MB
Setup Complex Easy Easy
Typo tolerance Plugin Built-in Built-in
Geo search Yes Yes Yes
Faceting Yes Yes Yes
Latency (p99) 50-200ms 20-50ms 5-50ms

Setup 

# Docker
docker run -p 8108:8108 \
  -v typesense-data:/data \
  typesense/typesense:27.0 \
  --data-dir=/data \
  --api-key=YOUR_API_KEY

# Or binary
curl -O https://dl.typesense.org/releases/27.0/typesense-server-27.0-linux-amd64.tar.gz
tar xzf typesense-server-*.tar.gz
./typesense-server --data-dir=./data --api-key=YOUR_API_KEY
Enter fullscreen mode Exit fullscreen mode

Create Collection & Index Data 

import Typesense from "typesense";

const client = new Typesense.Client({
  nodes: [{ host: "localhost", port: 8108, protocol: "http" }],
  apiKey: "YOUR_API_KEY",
});

// Create collection (schema)
await client.collections().create({
  name: "products",
  fields: [
    { name: "name", type: "string" },
    { name: "description", type: "string" },
    { name: "price", type: "float" },
    { name: "category", type: "string", facet: true },
    { name: "rating", type: "float" },
    { name: "in_stock", type: "bool" },
  ],
  default_sorting_field: "rating",
});

// Index documents
await client.collections("products").documents().import([
  { name: "MacBook Pro", description: "Powerful laptop for developers", price: 1999, category: "Laptops", rating: 4.8, in_stock: true },
  { name: "iPhone 15", description: "Latest smartphone from Apple", price: 999, category: "Phones", rating: 4.6, in_stock: true },
]);
Enter fullscreen mode Exit fullscreen mode

Search 

const results = await client.collections("products").documents().search({
  q: "macbok",              // Typo-tolerant!
  query_by: "name,description",
  filter_by: "price:<2000 && in_stock:true",
  sort_by: "rating:desc",
  facet_by: "category",
  per_page: 10,
});

// results.hits[0].document → { name: "MacBook Pro", ... }
// results.facet_counts → [{ field_name: "category", counts: [{value: "Laptops", count: 1}] }]
Enter fullscreen mode Exit fullscreen mode

Typo in "macbok"? Typesense still finds "MacBook Pro".

InstantSearch UI 

npm install typesense-instantsearch-adapter react-instantsearch
Enter fullscreen mode Exit fullscreen mode 
import TypesenseInstantsearchAdapter from "typesense-instantsearch-adapter";
import { InstantSearch, SearchBox, Hits, RefinementList } from "react-instantsearch";

const adapter = new TypesenseInstantsearchAdapter({
  server: { nodes: [{ host: "localhost", port: 8108, protocol: "http" }], apiKey: "search-key" },
  additionalSearchParameters: { query_by: "name,description" },
});

function SearchPage() {
  return (
    <InstantSearch indexName="products" searchClient={adapter.searchClient}>
      <SearchBox />
      <RefinementList attribute="category" />
      <Hits hitComponent={({ hit }) => <div>{hit.name} — ${hit.price}</div>} />
    </InstantSearch>
  );
}
Enter fullscreen mode Exit fullscreen mode

Full search UI with faceted filtering in 20 lines.

Cloud Option

Typesense Cloud starts at free (28K documents, 28K searches/month). No server to manage.

Need search or data extraction solutions? I build web scraping and API tools. Email spinov001@gmail.com or explore my Apify tools.

Top comments (0)