Back to Arena

Milvus

by Zilliz

System Card

OrganizationZilliz
Released2019-10
Architecturevector-rag / Distributed cloud-native vector DB
DetailsLarge-scale distributed vector database engineered for billion-scale vector search. Storage-compute separation, multiple ANN indexes, and a managed service via Zilliz Cloud.
Parameters—
Domainrag-retrieval
Open SourceYes
WebsiteVisit
CodeRepository
vector-dbdistributedbillion-scaleopen-source

Capability Profile

Benchmark Scores

5 of 14 benchmarks
Long-Context Retrieval
2/5
RULER
78.293p
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
72.860p
HotpotQA
52.65p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
73.487p
Sources:Milvus benchmarkMilvus (milvus-io/milvus); evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Milvus (milvus-io/milvus); evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Milvus (milvus-io/milvus); evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)