Migrating from Algolia to Meilisearch

This guide will take you step-by-step through the creation of a script to upload data indexed by Algolia to Meilisearch.

This page aims to help current users of Algolia make the transition to Meilisearch.

For a high-level comparison of the two search companies and their products, see our analysis of the search market.

Overview

This guide will take you step-by-step through the creation of a script to upload Algolia index data to Meilisearch. Examples are provided in JavaScript, Python, and Ruby. You can also skip directly to the finished script.

The migration process consists of three steps:

1. Export your data stored in Algolia
2. Import your data into Meilisearch
3. Configure your Meilisearch index settings (optional)

To help with the transition, we have also included a comparison of Meilisearch and Algolia’s API methods and front-end components.

Before continuing, make sure you have Meilisearch installed and have access to a command-line terminal. If you’re unsure how to install Meilisearch, see our quick start.

This guide includes examples in JavaScript, Python, and Ruby. The packages used:

• JavaScript: algoliasearch 4.x, meilisearch (compatible with Meilisearch v1.0+)
• Python: algoliasearch 4.x, meilisearch
• Ruby: algolia 3.x, meilisearch

Export your Algolia data

Initialize project

JavaScript

mkdir algolia-meilisearch-migration
cd algolia-meilisearch-migration
npm init -y
touch script.js

Install dependencies

JavaScript

npm install -s algoliasearch@4 meilisearch

Create Algolia client

You’ll need your Application ID and Admin API Key to start the Algolia client. Both can be found in your Algolia account.

Paste the below code in your script file:

JavaScript

const algoliaSearch = require("algoliasearch");

const algoliaClient = algoliaSearch(
  "APPLICATION_ID",
  "ADMIN_API_KEY"
);
const algoliaIndex = algoliaClient.initIndex("INDEX_NAME");

Replace APPLICATION_ID and ADMIN_API_KEY with your Algolia application ID and admin API key respectively.

Replace INDEX_NAME with the name of the Algolia index you would like to migrate to Meilisearch.

Fetch data from Algolia

To fetch all Algolia index data at once, use Algolia’s browseObjects method.

JavaScript

let records = [];
await algoliaIndex.browseObjects({
    batch: (hits) => {
      records = records.concat(hits);
    }
  });

The records array will contain all documents from your Algolia index. We will use records again later in the upload process.

Import your data into Meilisearch

Create Meilisearch client

Create a Meilisearch client by passing the host URL and API key of your Meilisearch instance. The easiest option is to use the automatically generated admin API key.

JavaScript

const { MeiliSearch } = require("meilisearch");

const meiliClient = new MeiliSearch({
  host: "MEILI_HOST",
  apiKey: "MEILI_API_KEY",
});
const meiliIndex = meiliClient.index("MEILI_INDEX_NAME");

Replace MEILI_HOST,MEILI_API_KEY, and MEILI_INDEX_NAME with your Meilisearch host URL, Meilisearch API key, and the index name where you would like to add documents. Meilisearch will create the index if it doesn’t already exist.

Upload data to Meilisearch

Next, use addDocumentsInBatches to upload all your records in batches of 100,000.

JavaScript

const BATCH_SIZE = 100000;
await meiliIndex.addDocumentsInBatches(records, BATCH_SIZE);

That’s all! When you’re ready to run the script, enter the below command:

JavaScript

node script.js

Finished script

JavaScript

const algoliaSearch = require("algoliasearch");
const { MeiliSearch } = require("meilisearch");

const BATCH_SIZE = 100000;

(async () => {
  const algoliaClient = algoliaSearch("APPLICATION_ID", "ADMIN_API_KEY");
  const algoliaIndex = algoliaClient.initIndex("INDEX_NAME");

  let records = [];
  await algoliaIndex.browseObjects({
    batch: (hits) => {
      records = records.concat(hits);
    }
  });

  const meiliClient = new MeiliSearch({
    host: "MEILI_HOST",
    apiKey: "MEILI_API_KEY",
  });
  const meiliIndex = meiliClient.index("MEILI_INDEX_NAME");

  await meiliIndex.addDocumentsInBatches(records, BATCH_SIZE);
})();

Configure your index settings

Meilisearch’s default settings are designed to deliver a fast and relevant search experience that works for most use-cases.

To customize your index settings, we recommend following this guide. To learn more about the differences between settings in Algolia and Meilisearch, read on.

Index settings vs. search parameters

One of the key usage differences between Algolia and Meilisearch is how they approach index settings and search parameters.

In Algolia, API parameters is a flexible category that includes both index settings and search parameters. Many API parameters can be used both at indexing time—to set default behavior—or at search time—to override that behavior.

In Meilisearch, index settings and search parameters are two distinct categories. Settings affect all searches on an index, while parameters affect the results of a single search.

Some Meilisearch parameters require index settings to be configured beforehand. For example, you must first configure the index setting sortableAttributes to use the search parameter sort. However, unlike in Algolia, an index setting can never be used as a parameter and vice versa.

Settings and parameters comparison

The below table compares Algolia’s API parameters with the equivalent Meilisearch setting or search parameter. The Type column indicates whether the Meilisearch equivalent is a search parameter (param), an index setting (setting), or both.

Query and pagination

Algolia	Meilisearch	Type
query	q	param
offset	offset	param
length	limit	param
page	page	param
hitsPerPage	hitsPerPage	param
paginatedHits limit	pagination.maxTotalHits	setting

Filtering and sorting

Algolia	Meilisearch	Type
filters	filter	param
facets	facets	param
attributesForFaceting	filterableAttributes	setting
maxValuesPerFacet	faceting.maxValuesPerFacet	setting
sortFacetValuesBy	faceting.sortFacetValuesBy	setting
maxFacetHits	facetSearch	setting
Sorting (using replicas)	sortableAttributes + sort param (no replicas required)	setting + param
distinct	distinct (per-query) or distinctAttribute (index-wide)	param + setting
attributeForDistinct	distinctAttribute	setting

Geo search

Algolia	Meilisearch	Type
aroundLatLng / aroundRadius	_geoRadius(lat, lng, radius) in filter	param
insideBoundingBox	_geoBoundingBox([lat, lng], [lat, lng]) in filter	param
insidePolygon	_geoPolygon([lat, lng], [lat, lng], ...) in filter	param
aroundPrecision	No direct equivalent	—

Highlighting and snippets

Algolia	Meilisearch	Type
attributesToHighlight	attributesToHighlight	param
highlightPreTag	highlightPreTag	param
highlightPostTag	highlightPostTag	param
attributesToSnippet	attributesToCrop + cropLength	param
snippetEllipsisText	cropMarker	param
restrictHighlightAndSnippetArrays	Not supported	—

Attributes and ranking

Algolia	Meilisearch	Type
searchableAttributes	searchableAttributes	setting
restrictSearchableAttributes	attributesToSearchOn	param
attributesToRetrieve	attributesToRetrieve	param
unretrievableAttributes	No direct equivalent; achieved by removing attributes from displayedAttributes	setting
ranking	rankingRules	setting
customRanking	Integrated within rankingRules	setting
getRankingInfo	showRankingScore / showRankingScoreDetails	param

Typo tolerance and language

Algolia	Meilisearch	Type
typoTolerance	typoTolerance	setting
disableTypoToleranceOnAttributes	typoTolerance.disableOnAttributes	setting
removeStopWords	stopWords	setting
synonyms	synonyms	setting
separatorsToIndex	separatorTokens / nonSeparatorTokens	setting
naturalLanguages	localizedAttributes setting + locales param	setting + param
queryType (prefix matching)	prefixSearch	setting
removeWordsIfNoResults	Automatically supported via matchingStrategy param	param

AI and vector search

Algolia	Meilisearch	Type
NeuralSearch mode	hybrid param + embedders setting	setting + param
enableReRanking	Integrated in hybrid search	param
AI personalization	personalize	param

Other

Algolia	Meilisearch	Type
analytics / clickAnalytics	Separate Analytics API	—
disablePrefixOnAttributes	Not supported	—
relevancyStrictness	rankingScoreThreshold	param

API methods

This section compares Algolia and Meilisearch’s respective API methods, using JavaScript for reference.

Method	Algolia	Meilisearch
Index Instantiation	client.initIndex() Here, client is an Algolia instance.	client.index() Here, client is a Meilisearch instance.
Create Index	Algolia automatically creates an index the first time you add a record or settings.	The same applies to Meilisearch, but users can also create an index explicitly: client.createIndex(string indexName)
Get All Indexes	client.listIndices()	client.getIndexes()
Get Single Index	No method available	client.getIndex(string indexName)
Delete Index	index.delete()	client.deleteIndex(string indexName)
Get Index Settings	index.getSettings()	index.getSettings()
Update Index Settings	index.setSettings(object settings)	index.updateSettings(object settings)
Search Method	index.search(string query, { searchParameters, requestOptions })	index.search(string query, object searchParameters)
Add Object	index.saveObjects(array objects)	index.addDocuments(array objects)
Partial Update Object	index.partialUpdateObjects(array objects)	index.updateDocuments(array objects)
Delete All Objects	index.deleteObjects(array objectIDs)	index.deleteAllDocuments()
Delete One Object	index.deleteObject(string objectID)	index.deleteDocument(string id)
Get All Objects	index.getObjects(array objectIDs)	index.getDocuments(object params)
Get Single Object	index.getObject(str objectID)	index.getDocument(string id)
Get API Keys	client.listApiKeys()	client.getKeys()
Get API Key Info	client.getApiKey(string apiKey)	client.getKey(string apiKey)
Create API Key	client.addApiKey(array acl)	client.createKey(object configuration)
Update API Key	client.updateApiKey(string apiKey, object configuration)	client.updateKey(string apiKey, object configuration)
Delete API Key	client.deleteApiKey(string apiKey)	client.deleteKey(string apiKey)

Front-end components

InstantSearch is a collection of open-source tools maintained by Algolia and used to generate front-end search UI components. To use InstantSearch with Meilisearch, you must use Instant Meilisearch.

Instant Meilisearch is a plugin connecting your Meilisearch instance with InstantSearch, giving you access to many of the same front-end components as Algolia users. You can find an up-to-date list of the components supported by Instant Meilisearch in the GitHub project’s README.