Back to Arena

Manticore Search

by Manticore Software Ltd.

System Card

OrganizationManticore Software Ltd.
Released2017-01
Architecturevector-rag / open-source SQL search engine with vector
DetailsManticore Search is an open-source full-text search engine (Sphinx Search fork) that added HNSW-based vector search in version 6.3.0 (May 2024). It uses SQL as its native query language and is compatible with MySQL protocol. Combines full-text, faceted, geo-spatial, and vector search in one engine with real-time indexing. Acts as a drop-in Elasticsearch alternative.
Parameters—
Domainrag-retrieval
Open SourceYes
WebsiteVisit
full-textSQLElasticsearch-alternativeHNSWreal-time-indexing

Capability Profile

Benchmark Scores

5 of 14 benchmarks
Long-Context Retrieval
2/5
RULER
66.410p
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
62.310p
HotpotQA
62.823p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
64.922p
Sources:Manticore Search vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Manticore Search vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Manticore Search vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Manticore Search vendor documentation; evaluated on RAGAS: Automated Evaluation of Retrieval-Augmented Generation (Exploding Gradients, 2309)Manticore Search vendor documentation; evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)