Back to Arena

LlamaIndex Memory

by LlamaIndex

System Card

OrganizationLlamaIndex
Released2023-11
Architecturehybrid / Vector + chat memory buffers
DetailsMemory module for the LlamaIndex framework offering chat buffers, vector memory, and composable memory blocks. Plugs into any LlamaIndex agent or query engine.
Parameters—
Domainrag-retrievalagent-memoryepisodic-session
Open SourceYes
WebsiteVisit
CodeRepository
llamaindexframeworkbuffersmodular

Capability Profile

Benchmark Scores

7 of 14 benchmarks
Long-Context Retrieval
1/5
RULER
no data
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
1/2
LoCoMo
56.56p
MemoryBank
no data
Cross-Session Memory
1/1
LongMemEval
56.83p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
70.236p
HotpotQA
67.537p
Agent Task Memory
1/1
AgentBench-Mem
59.74p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
72.173p
Sources:LlamaIndex evalsLlamaIndex Memory (run-llama/llama_index); evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)LlamaIndex Memory (run-llama/llama_index); evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)LlamaIndex Memory (run-llama/llama_index); evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)LlamaIndex Memory (run-llama/llama_index); evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)LlamaIndex Memory (run-llama/llama_index); evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)