manticoresoftware/manticoresearch

Health v2 · maintenance only

Adoption not part of the score

11956 stars · 641 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Manticore Search is an open-source search database written in C++ that provides fast full-text, vector, and hybrid search with real-time indexing. It supports SQL and JSON APIs over the MySQL protocol and HTTP, positioning itself as a lightweight, high-performance alternative to Elasticsearch.

Use cases

• add full-text search to my website or app
• run semantic and vector similarity search over embeddings
• self-host an Elasticsearch alternative with lower resource usage
• power product catalog and faceted search
• build autocomplete and fuzzy search suggestions
• do log analytics over millions of records
• perform geo and location-based search
• query large datasets with SQL over the MySQL protocol

When to choose

• you need fast full-text and vector search with low latency and low memory footprint
• you want SQL/JSON query interfaces and MySQL client compatibility
• you need real-time indexing, replication, and sharding in a self-hosted search engine
• Elasticsearch is too heavy or costly for your infrastructure

When to avoid

• you need the broader Elastic ecosystem (Kibana, ML jobs, managed cloud)
• you require a permissively licensed database (Manticore is GPL-3.0)
• your workload is purely relational OLTP without search requirements

Facets

service · maturity stable

search-engine database vector-database http-server api-framework databases big-data analytics windows self-hosted cpp full-text-search bm25 elasticsearch-alternative hybrid-search mysql-protocol real-time-indexing sphinx-fork sql search natural-language-processing linux macos docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="manticoresoftware/manticoresearch")