Back to Arena

Mnemosyne

by independent

System Card

Organizationindependent
Released2025-10
Architecturegraph-rag / Human-inspired probabilistic recall with decay
DetailsUnsupervised long-term memory for edge LLMs using graph-structured storage, modular substance/redundancy filters, memory pruning, and probabilistic recall with temporal decay modeled on human memory. Includes a "core summary" of personality/domain details.
Parameters—
Domainpersonalizationepisodic-sessionagent-memory
Open SourceNo
PaperView Paper
edge-llmdecaygraphunsupervisedpersonalized

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Data Transparency:1 self-reported5 estimated
Long-Context Retrieval
0/5
RULER
no data
NIAH
no data
LooGLE
no data
LongBench
no data
∞Bench
no data
Multi-Turn Recall
2/2
LoCoMo
54.51pSelf-Reported
MemoryBank
64.413pEstimated
Cross-Session Memory
1/1
LongMemEval
59.912pEstimated
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
66.128pEstimated
HotpotQA
64.829pEstimated
Agent Task Memory
1/1
AgentBench-Mem
617pEstimated
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:arXiv:2510.08601 Table 2 — Overall J-score, Llama3.1-8B-Instruct; SingleHop 62.78, MultiHop 49.53, OpenDomain 60.42, Temporal 53.03Arena estimate — derived from capability profile, not independently verified