Elasticsearch / OpenSearch
Mature hybrid (BM25 + vector) implementations. RRF + weighted combination. Elastic ELSER sparse model.
At a glance
- Type
- Mature hybrid search
- Tier
- T1
- Section
- Vector-database infrastructure
- Created
- 2012
- Latest release
- not applicable — not OSS
- License
- not applicable — not OSS
- GitHub
- not applicable — no GitHub repo
- Pricing
- Free + paid
- Funding
- $162M pre-IPO; IPO 2018 NYSE:ESTC; public company
Taxonomy
- storage
- vector
- retrieval
- similarity
- persistence
- long-term
- update
- overwrite
- unit
- document
- governance
- inspectable
- conflict
- overwrite
When to use
Optimised for: low-latency similarity search + scale
Anti-fit: not for relational / graph-heavy queries (vector-first by design)
Pros & cons
Pros
Ubiquitous in enterprise stacks — vector search added without requiring a new database; mature ops, observability, and security.
Cons
Vector performance is behind dedicated engines; ML extensions vary across the Elastic vs OpenSearch fork.
Claims & capabilities
Elastic positions Elasticsearch as world's most-downloaded vector DB; Elasticsearch reports 2x–12x faster than OpenSearch on its own benchmarks; independent Trail of Bits study (March 2025) found OpenSearch 1.6x faster on Big5 mixed workload and 11% faster on vector. Customers include Reed (UK), Stack Overflow, Adobe; OpenSearch on AWS recommended for Bedrock RAG
Technical surface
- API surface
- REST, SDK: many
- Backend storage
- custom (Lucene-backed)
- Deployment
- Both
- Embedding model
- multiple supported
- Multi-tenancy
- Allocator-level cluster separation in Elastic Cloud Enterprise; index-level access control; dedicated single-tenant deployments available
- MCP
- via official adapter — Elastic MCP server
- A2A
- no Google A2A (Agent2Agent) integration documented as of 2026-05.
- OpenTelemetry
- first-class — native OTel
Similar systems
Other vector-database infrastructure in the catalog, ranked by inbound references.
- Qdrant T1
Distribution-Based Score Fusion + RRF. Sparse vectors native; filtering via ANN graph modification.
- pgvector T1
Stores embeddings alongside relational + full-text data. HNSW + IVFFlat ANN indexes. Used as agent conversation memory via LangChain + MCP. Foundation of Supabase AI and many self-hosted RAG stacks.
- Pinecone T1
Managed vector DB. Cascading sparse + dense + rerank pipeline; pinecone-rerank-v0 .
- Chroma T1
Limited native hybrid (users build RRF custom). Fast Rust core (v2.5).
- LanceDB T1
Embedded vector DB (Arrow columnar). RRF reranker. Petabyte-scale on disk.
- MongoDB Atlas Vector Search T1
Agent memory store for both short-term (document) and long-term (vector). LangGraph checkpointer for stateful agents. Vector search extended to Community Edition (Sept 2025).