Meilisearch Search Engine

Meilisearch is a free and open-source search engine that provides fast, relevant, and typo-tolerant full-text search out of the box with zero configuration. Developed by Meilisearch (formerly Paris-based Meili, founded in 2018 by Clmentine Urkizar-Madina and Quentin de Quelen), Meilisearch is written in Rust and designed for instant search experiences (sub-50 millisecond response times) on datasets of any size. Key features: instant search: Meilisearch returns search results in under 50 milliseconds even on datasets with millions of documents, enabling as-you-type search experiences. Typo tolerance: automatically handles typos and misspellings, finding relevant results even when the search query contains errors. The typo tolerance is configurable (minimum word length for typo, disabled attributes). Relevance ranking: a built-in relevance ranking algorithm (the Bucket Sort) combines multiple ranking rules: exactness, word position, attribute, words, typo, proximity, and custom ranking rules. Rules can be reordered and customized per index. Multi-language support: optimized tokenizer for Latin-based languages, Chinese, Japanese, Korean, Thai, and Hebrew, with language detection and proper word segmentation. Faceted search: filter and group results by categories (facets), enabling navigation and filtering interfaces. Synonyms: define synonyms for terms to expand search coverage. Stop words: exclude common words from indexing and search to improve relevance. Geosearch: filter and sort results by geographic distance using latitude/longitude data. Indexing: documents are indexed in the background, allowing the search engine to continue serving queries during reindexing. RESTful API: simple HTTP/JSON API for all operations (create index, add documents, search). SDKs for JavaScript, Python, PHP, Ruby, Go, Rust, Java, Swift, .NET. Cross-platform. MIT.