Meilisearch 1.15
Meilisearch 1.15 introduces a new typo tolerance setting, comparison operators for string filters, and improved support for Chinese.

We're happy to announce the release of Meilisearch v1.15. In this article, we'll highlight the key changes and improvements in this release.
For a complete list of all updates and fixes, please visit the changelog on GitHub.
These powerful new features are available in Meilisearch Cloud. Sign up now to be among the first to experience the latest improvements!
New: Disable typo tolerance for numbers
With Meilisearch v1.15, you can turn off typo tolerance for numeric values—just like you already do for words and attributes.
A search for 2024
will match only 2024
, not 2025
or 2004
. This is especially helpful when searching for postal addresses or phone numbers.
Skipping number fuzziness can also speed up your indexing, especially for datasets that contain many unique numbers.
New: Lexicographic string filters
You can now compare any string with <
, <=
, >
, >=
or an open range TO
—just like numbers—using lexicographic order.
This is ideal for scenarios where you need to filter by human-readable values—such as dates (e.g., “2025-05-28”) instead of numeric timestamps.
The following filter finds every record released before July 17, 2023:
release_date < "2023-07-17"
Improvement: better support for Chinese
We’ve enhanced the Charabia tokenizer to provide more precise segmentation of Chinese text. If your dataset contains Chinese-language content, you must reindex your database—otherwise, Chinese queries and documents will not be processed correctly and may be ignored during search.
To reindex, you can use a dump or—if you are using the dumpless upgrade—trigger a reindexing with one of the following methods:
-
Add Chinese to your localized attributes Include the Chinese locale (e.g., cmn) in your localizedAttributes.
-
Update your stop-word list or dictionary Make any change to your stop words or custom dictionary entries.
-
Refresh the searchable attribute Temporarily remove an attribute from searchable attributes list, then re-add it once the removal task has completed.
Contributors shout-out
We want to give a massive thank you to the contributors who made this release possible. Special thanks to CodeMan62, shaokeyibb, HDT3213, shu-kitamura, DanasFi, vuthanhtung2412, and ZeroZ-lab for their contributions to Meilisearch; to luflow, mosuka, and HDT3213 for their work on Charabia; and to lucascool12 and Dr-Emann for their efforts on RoaringBitmap.
And that’s a wrap for v1.15! These release notes only highlight the most significant updates. For an exhaustive listing, read the changelog on GitHub.
For more information, subscribe to our monthly newsletter, or join our Product Discussions.
For anything else, join our developers community on Discord.