Back to Arena

Nomi AI

by Glimpse AI, Inc.

System Card

OrganizationGlimpse AI, Inc.
Released2023-01
Architecturehierarchical-summary / Identity Core + Multi-Tier Persistent Memory
DetailsNomi uses a short/medium/infinite long-term memory stack augmented by an Identity Core (launched December 2024) — a living memory core that dynamically evolves the companion's self-concept based on accumulated interactions. An interactive Mind Map 2.0 visualizes how the companion connects different memories, concepts, and knowledge into a growing relational graph. Ranked highest in independent memory recall tests, capturing 23/25 specific facts weeks after they were shared.
Parameters—
Domainpersonalizationepisodic-sessionlifelong-learning
Open SourceNo
WebsiteVisit
companionidentity-coremulti-tier-memorymind-mapemotional-aibest-in-class-recall

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
2/2
LoCoMo
79.986p
MemoryBank
78.873p
Cross-Session Memory
1/1
LongMemEval
76.655p
Multi-Hop QA
1/3
BABILong
7658p
MultiHop-RAG
no data
HotpotQA
no data
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
1/1
PerLTQA
75.741p
Factuality / Grounding
0/1
RAGAS
no data
Sources:Nomi AI vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Nomi AI vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Nomi AI vendor documentation; evaluated on MemoryBank: Enhancing LLMs with Long-Term Memory (Sun Yat-sen University, 2305)Nomi AI vendor documentation; evaluated on PerLTQA: A Personal Long-Term Memory Question Answering Dataset (PolyU, 2402)Nomi AI vendor documentation; evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)Nomi AI vendor documentation; evaluated on BABILong: Testing the Limits of LLMs with Long-Context Reasoning-in-a-Haystack (AIRI, 2406)