Back to Arena

Cognigy

by Cognigy GmbH (acquired by NICE, July 2025)

System Card

OrganizationCognigy GmbH (acquired by NICE, July 2025)
Released2016-01
Architectureagentic-workflow / enterprise conversational AI with persistent memory
DetailsCognigy.AI is an enterprise-grade conversational AI platform for CX (contact center automation), handling voice and chat at scale. Agents maintain persistent memory of customer preferences and past interactions across sessions, with a knowledge graph for enterprise data access. Targeted Knowledge Retrieval (2025) allows agents to query specific data segments rather than full corpora. Powers 1B+ annual interactions.
Parameters—
Domainagent-memorypersonalizationepisodic-session
Open SourceNo
WebsiteVisit
enterprise-CXvoice-AIpersistent-memoryknowledge-retrievalcontact-center

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
80.990p
MemoryBank
78.672p
Cross-Session Memory
1/1
LongMemEval
82.190p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
74.270p
HotpotQA
75.775p
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:Cognigy vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Cognigy vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Cognigy vendor documentation; evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)Cognigy vendor documentation; evaluated on MemoryBank: Enhancing LLMs with Long-Term Memory (Sun Yat-sen University, 2305)Cognigy vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Cognigy vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)