Back to Arena

Mendable

by Mendable (YC-backed)

System Card

OrganizationMendable (YC-backed)
Released2022-10
Architecturevector-rag / Knowledge-Base Chat RAG
DetailsMendable ingests technical documentation through 20+ connectors (GitHub, Notion, Confluence, Google Drive, Zendesk) and exposes a retrieval-augmented chat layer that companies embed for customer-facing or internal support. It supports multiple base LLMs including GPT-3.5-Turbo and GPT-4, with continuous learning via human feedback that updates the retrieval index in real time. Enterprise tier includes SSO, RBAC, and SOC 2 Type II.
Parameters—
Domainrag-retrieval
Open SourceNo
WebsiteVisit
documentation-chatycragenterprisecustomer-supportfeedback-loop

Capability Profile

Benchmark Scores

5 of 14 benchmarks
Long-Context Retrieval
2/5
RULER
64.51p
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
0/2
LoCoMo
no data
MemoryBank
no data
Cross-Session Memory
0/1
LongMemEval
no data
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
62.711p
HotpotQA
64.629p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
63.77p
Sources:Mendable vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Mendable vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Mendable vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Mendable vendor documentation; evaluated on RAGAS: Automated Evaluation of Retrieval-Augmented Generation (Exploding Gradients, 2309)Mendable vendor documentation; evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)