Elasticsearch
Mature hybrid (BM25 + vector) implementations. RRF + weighted combination. Elastic ELSER sparse model.
Active · Verified 2026-10-26 · fresh as of 2026-10-26 · 55 cited facts, 5 with a third-party source · 2 recorded changes · Suggest a correction
At a glance
- Type
- Mature hybrid search vendor fix
- Tier
- T1
- Section
- Vector-database infrastructure
- Latest release
- not applicable — managed service repo fix
- License
- AGPL-3.0 / SSPL-1.0 / Elastic-2.0 (choice) repo fix
- Funding
- $162M pre-IPO; IPO 2018 NYSE:ESTC; public company third-party fix
- Freshness
- 2026-10-26 fix
- Layer
- model-infra repo fix
- SDK languages
- csharp, go, java, php, python, rest, ruby, rust, typescript vendor fix
- Protocols
- a2a, mcp-client, mcp-server, opentelemetry vendor fix
- Compliance
- csa-star, fedramp, hipaa, iso27001, iso27017, iso27018, pci-dss, soc3 vendor fix
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.
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
- yes — official Elastic Agent Builder MCP endpoint (Elastic 9.2+ / Serverless); older elastic/mcp-server-elasticsearch deprecated
- A2A
- no Google A2A (Agent2Agent) integration documented as of 2026-05.
- OpenTelemetry
- first-class — native OTel
Alternatives to Elasticsearch → live options ranked by shared design, licence, hosting and MCP support
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.
- Chroma T1
Limited native hybrid (users build RRF custom). Fast Rust core (v2.5). Acquired by Replit (definitive agreement announced 2026-10-23); Chroma Cloud is winding down (writes disabled 2026-11-22, reads disabled 2026-12-22) while open-source ChromaDB continues under Apache-2.0.
- 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.
- Weaviate T1
Hybrid Search 2.0 with learned ranking. BM25 + vector in unified index.
- Milvus T1
Multi-vector columns (10 simultaneous). Native hybrid search (v2.5). CAGRA + Vamana GPU/CPU (v2.6).
- Pinecone T1
Managed vector DB. Cascading sparse + dense + rerank pipeline; pinecone-rerank-v0 .
Sources (54 cited facts, 5 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Mature hybrid search vendor
- What it does & what's distinct Mature hybrid (BM25 + vector) implementations. RRF + weighted combination. Elastic ELSER… vendor
- Claims / benefits Elastic positions Elasticsearch as world's most-downloaded vector DB; Elasticsearch repo… vendor
- Modalities text vendor
- Created 2012 third-party
- Latest release not applicable — managed service repo
- License AGPL-3.0 / SSPL-1.0 / Elastic-2.0 (choice) repo
- GitHub elastic/elasticsearch — ~78k stars, active (commit date not captured) repo
- Mindshare One of the highest-mindshare entries in the table — Elasticsearch dominates general-purp… third-party
- Funding $162M pre-IPO; IPO 2018 NYSE:ESTC; public company third-party
- Customers / scale Thousands of enterprises; exact count not disclosed vendor
- Pricing free (OSS) self-managed; Elastic Cloud paid (price pages not re-verified) repo
- Compliance SOC 2 Type II; ISO 27001; HIPAA; FedRAMP (Elastic Cloud Gov); GDPR-compliant vendor
- Data handling Customer data not used for training; ISO 27701 / 27018 certified; GDPR compliant; all cl… vendor
- Deployment Both vendor
- HQ US third-party
- Founders / pedigree Shay Banon (CTO/co-founder), Simon Willnauer, Steven Schuurman, Uri Boness; founded 2012 third-party
- Memory volume / scale no scale claim (petabytes in enterprise use) vendor
- Performance Elastic 8.12 + RaBitQ binary quantisation: 32× memory reduction, recall@10 >0.95 on MS M… vendor
- API surface REST, SDK: many vendor
- Backend storage custom (Lucene-backed) vendor
- Multi-tenancy Allocator-level cluster separation in Elastic Cloud Enterprise; index-level access contr… vendor
- Encryption AES-256 at rest, TLS 1.2 in transit; BYOK via AWS KMS / Azure Key Vault / GCP Cloud KMS … vendor
- SSO / RBAC SSO + RBAC vendor
- Embedding model multiple supported vendor
- Versioning immutable append-log vendor
- Schema evolution schema with migration vendor
- Namespace primitives per-tenant vendor
- Contradiction handling always overwrite (last-write-wins) vendor
- Forgetting policy none (permanent) vendor
- MCP support yes — official Elastic Agent Builder MCP endpoint (Elastic 9.2+ / Serverless); older ela… repo
- OTel first-class — native OTel vendor
- Webhooks / events webhook + alerting actions vendor
- Import / export JSON (bulk API), snapshots, CSV vendor
- Validated verticals RAG developers, agent builders (broad horizontal); financial-services, healthcare, legal… vendor
- Time to running <10min (managed cloud) to <1hr (self-host Docker) vendor
- Anti-fit not for relational / graph-heavy queries (vector-first by design) vendor
- Optimised for low-latency similarity search + scale vendor
- Adjacent infrastructure BYO embedding model + LLM; this IS the vector store vendor
- Pros Ubiquitous in enterprise stacks — vector search added without requiring a new database; … vendor
- Cons Vector performance is behind dedicated engines. vendor
- Project & sources elastic.co vendor
- cost-input-usd-per-mtok 0 vendor
- cost-output-usd-per-mtok 0 vendor
- cost-tier free vendor
- cost-pricing-model free-tier vendor
- cost-last-verified 2026-06-29 vendor
- architecture-deployment-locus hybrid vendor
- licensing-bucket open-core repo
- positioning-platform-vs-product platform vendor
- taxonomy-three-layer-rollup model-infra repo
- compliance-certifications csa-star,fedramp,hipaa,iso27001,iso27017,iso27018,pci-dss,soc3 vendor
- sdk-languages csharp,go,java,php,python,rest,ruby,rust,typescript vendor
- protocol-support a2a,mcp-client,mcp-server,opentelemetry vendor
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-27
- Name “Elasticsearch / OpenSearch” → “Elasticsearch”
- License “AGPL-3.0 / SSPL-1.0 / Elastic-2.0 (choice) for Elasticsearch; OpenSearch fork is Apache-2.0” → “AGPL-3.0 / SSPL-1.0 / Elastic-2.0 (choice)” source
- Latest release “not applicable — not OSS” → “not applicable — managed service” source
- Pricing “free (OSS) self-managed; Elastic Cloud / Amazon OpenSearch Service paid (price pages not re-verified)” → “free (OSS) self-managed; Elastic Cloud paid (price pages not re-verified)” source
- 2026-10-26
- License “not applicable — not OSS” → “AGPL-3.0 / SSPL-1.0 / Elastic-2.0 (choice) for Elasticsearch; OpenSearch fork is Apache-2.0” source
- Pricing “Free + paid” → “free (OSS) self-managed; Elastic Cloud / Amazon OpenSearch Service paid (price pages not re-verified)” source
- MCP support “via official adapter — Elastic MCP server” → “yes — official Elastic Agent Builder MCP endpoint (Elastic 9.2+ / Serverless); older elastic/mcp-server-elasticsearch deprecated” source
Related systems
References (7)
- Anthropic Claude (foundation models) integrates with — Elastic Search Labs has a Anthropic Claude (foundation models) integration page. (verified 2026-10-29)
- Cohere Command R+ / Command A integrates with — Elastic Search Labs has a Cohere Command R+ / Command A integration page. (verified 2026-10-29)
- Jina AI Embeddings integrates with — Elastic Search Labs has a Jina AI Embeddings integration page. (verified 2026-10-29)
- LangChain (framework) integrates with — Elastic Search Labs has a LangChain (framework) integration page. (verified 2026-10-29)
- LlamaIndex (framework) integrates with — Elastic Search Labs has a LlamaIndex (framework) integration page. (verified 2026-10-29)
- Mistral Large 2 / Mixtral family integrates with — Elastic Search Labs has a Mistral Large 2 / Mixtral family integration page. (verified 2026-10-29)
- OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) integrates with — Elastic Search Labs has a OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) integration page. (verified 2026-10-29)
Referenced by (13)
- Cohere Embed integrates with — Cohere docs "Integrating Embedding Models with Other Tools": Elasticsearch and Cohere guide. (verified 2026-10-29)
- Dify integrates with — Dify self-host VECTOR_STORE supports `elasticsearch` as the knowledge-base vector store. (verified 2026-10-29)
- Haystack (framework) integrates with — Haystack integrations directory has an Elasticsearch integration page ("Use an Elasticsearch database with Haystack"). (verified 2026-10-29)
- LangChain (framework) integrates with — LangChain docs have an Elasticsearch integration provider page (Distributed search and analytics engine). (verified 2026-10-29)
- Langflow integrates with — Langflow components bundle: Elasticsearch vector store component. (verified 2026-10-29)
- LlamaIndex (framework) integrates with — LlamaIndex has an official vector store integration package llama-index-vector-stores-elasticsearch in its repo. (verified 2026-10-29)
- Mastra integrates with — Mastra repo ships the @mastra/elasticsearch store package. (verified 2026-10-29)
- Mem0 integrates with — Mem0 docs have an Elasticsearch vector store page for Mem0 open source. (verified 2026-10-29)
- Microsoft Semantic Kernel integrates with — Semantic Kernel docs list an out-of-the-box Elasticsearch vector store connector. (verified 2026-10-29)
- OpenLLMetry (Traceloop) integrates with — OpenLLMetry's What's Supported page lists Elasticsearch as instrumented out of the box. (verified 2026-10-29)
- OpenRouter integrates with — OpenRouter Broadcast docs have a page for sending request traces to Elastic Observability. (verified 2026-10-29)
- OpenSearch forks — AWS-led fork of Elasticsearch 7 — Apache-licensed
- Phidata (now Agno) integrates with — Agno ships a elasticsearch vector DB integration module (agno.vectordb.elasticsearch). (verified 2026-10-29)