Back to Arena

Lindy AI

by Lindy AI

System Card

OrganizationLindy AI
Released2023-01
Architectureagentic-workflow / Personalized AI Chief-of-Staff
DetailsLindy builds persistent preference memories from user feedback across email, calendar, and CRM interactions, adapting its writing style and scheduling behavior over time. Lindy 3.0 introduced "Agentic Reasoning" enabling browser navigation and multi-app task completion with self-correction. Memory is accumulated across sessions to continuously adjust to user priorities, style, and workflow patterns.
Parameters—
Domainagent-memorypersonalizationepisodic-session
Open SourceNo
WebsiteVisit
productivityemailcalendarworkflow-automationpersonalizationagentic

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
7439p
MemoryBank
81.586p
Cross-Session Memory
1/1
LongMemEval
80.583p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
72.759p
HotpotQA
72.859p
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:Lindy AI vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Lindy AI vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Lindy AI vendor documentation; evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)Lindy AI vendor documentation; evaluated on MemoryBank: Enhancing LLMs with Long-Term Memory (Sun Yat-sen University, 2305)Lindy AI vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Lindy AI vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)