Manticore Search

Manticore Search is an open-source search database that combines advanced full-text search, vector and hybrid search, real-time indexing, and analytical queries in one lightweight C++ engine. It supports SQL and JSON APIs for querying and can scale from a single server to distributed deployments.

Manticore Search is 2.83x faster than Elasticsearch for big data (1.7 billion documents) and 10.09x faster for log analytics (10M Nginx logs). It offers better performance with lower resource consumption and serves as a compatible alternative that can seamlessly replace Elasticsearch while integrating with tools like Logstash, Beats, and Kibana.

Manticore Search supports full-text search with typo tolerance, vector and semantic search, hybrid search combining words and meaning, conversational search powered by LLM providers, geo-spatial search, and image search. It also provides fuzzy matching, faceting, query suggestions, and autocomplete capabilities.

Yes, Manticore Search is 100% open source and free to use. It is published under OSI-approved open source licenses. While the core product is free, the Manticore team offers optional professional services including consulting, support, feature development, and custom implementations through Community, Standard, and Enterprise subscription plans.