OpenSearch is an open-source search and analytics suite, licensed under Apache 2.0, that can also store embeddings and run vector, semantic and hybrid search for AI apps.
In plain words
OpenSearch is a distributed search engine with a web dashboard for exploring data. It includes some Apache-licensed code that came from Elasticsearch, and it is a community project hosted by LF Projects. Besides keyword search and log analysis, it can store vector embeddings next to your other data and search them by similarity, and it can generate embeddings for you. It supports semantic, hybrid, multimodal and sparse search.
A quick example
Imagine a company that already keeps its product catalogue in OpenSearch for keyword search. It creates a vector index and lets OpenSearch generate embeddings for each product description. A shopper's plain-language question can then be matched by meaning, by keywords, or by a hybrid of both.
Where you'll meet it
- Semantic search and retrieval for RAG.
- Recommendation systems.
- Log analytics and observability dashboards.
Related
Sources
- Docsopensearch-project/OpenSearch, OpenSearch Project (GitHub) · read 29 Sept 2026
- DocsAbout OpenSearch, OpenSearch documentation · read 29 Sept 2026
- DocsVector search, OpenSearch documentation · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.