Back to Arena
Neon Vector
by Neon Inc.
System Card
OrganizationNeon Inc.
Released2022-04
Architecturevector-rag / serverless Postgres with pgvector
DetailsNeon is a serverless PostgreSQL cloud service that supports pgvector natively, enabling vector similarity search with auto-scaling compute and storage. Its branching feature lets developers create database branches for testing RAG pipelines before deploying to production. Raised $104M in total funding. Microsoft Semantic Kernel has a native Neon connector.
Parameters—
Domainrag-retrieval
Open SourcePartial
WebsiteVisit
serverlessbranchingpgvectorPostgreSQLscale-to-zero
Capability Profile
Benchmark Scores
5 of 14 benchmarksMulti-Turn Recall0/2
LoCoMo
no dataMemoryBank
no dataCross-Session Memory0/1
LongMemEval
no dataMulti-Hop QA2/3
Agent Task Memory0/1
AgentBench-Mem
no dataPersonalization0/1
PerLTQA
no dataFactuality / Grounding1/1
Sources:Neon Vector vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Neon Vector vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Neon Vector vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Neon Vector vendor documentation; evaluated on RAGAS: Automated Evaluation of Retrieval-Augmented Generation (Exploding Gradients, 2309)Neon Vector vendor documentation; evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)