Back to Arena

Elasticsearch Vector

by Elastic N.V.

System Card

OrganizationElastic N.V.
Released2019-07
Architecturevector-rag / inverted-index-plus-dense-vector
DetailsElasticsearch added dense_vector field type and script score queries in 7.3 (2019), then HNSW ANN indexing in 8.0 (2022), making it a hybrid full-text + semantic search engine. Supports RRF (Reciprocal Rank Fusion) for combining lexical and semantic result sets. Integrated into Elastic Cloud and the Elastic AI Assistant for RAG over enterprise corpora.
Parameters—
Domainrag-retrieval
Open SourcePartial
WebsiteVisit
hnswfull-textrrfelastic-cloudenterprise

Capability Profile

Benchmark Scores

5 of 14 benchmarks
Long-Context Retrieval
2/5
RULER
69.437p
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
0/2
LoCoMo
no data
MemoryBank
no data
Cross-Session Memory
0/1
LongMemEval
no data
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
65.425p
HotpotQA
62.823p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
64.213p
Sources:Elasticsearch Vector vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Elasticsearch Vector vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Elasticsearch Vector vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Elasticsearch Vector vendor documentation; evaluated on RAGAS: Automated Evaluation of Retrieval-Augmented Generation (Exploding Gradients, 2309)Elasticsearch Vector vendor documentation; evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)