Memary

by Kingjulio8238

System Card

OrganizationKingjulio8238

Released2024-04

Architecturegraph-rag / Knowledge graph + Llama-Index

DetailsBuilds a Neo4j knowledge graph of entities and relations extracted from agent traces, paired with a Llama-Index retriever. Designed as a drop-in long-term memory layer for autonomous agents.

Parameters—

Domainagent-memoryknowledge-graphlifelong-learning

Open SourceYes

CodeRepository

neo4jllama-indexagentopen-source

Capability Profile

Benchmark Scores

6 of 14 benchmarks

Long-Context Retrieval

0/5

RULER

no data

NIAH

no data

LooGLE

no data

LongBench

no data

∞Bench

no data

Multi-Turn Recall

1/2

LoCoMo

71.728p

MemoryBank

no data

Cross-Session Memory

1/1

LongMemEval

75.550p

Multi-Hop QA

3/3

BABILong

76.864p

MultiHop-RAG

52.70p

HotpotQA

54.26p

Agent Task Memory

1/1

AgentBench-Mem

7226p

Personalization

0/1

PerLTQA

no data

Factuality / Grounding

0/1

RAGAS

no data

Sources:Memary repo evals Memary (kingjulio8238/Memary); evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Memary (kingjulio8238/Memary); evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Memary (kingjulio8238/Memary); evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)Memary (kingjulio8238/Memary); evaluated on BABILong: Testing the Limits of LLMs with Long-Context Reasoning-in-a-Haystack (AIRI, 2406)