Back to Arena

Ragie

by Ragie Inc.

System Card

OrganizationRagie Inc.
Released2024-08
Architecturevector-rag / fully managed RAG-as-a-service
DetailsRagie (founded by Bob Remeika and Mohammed Rafiq, both ex-enterprise software veterans) provides a fully managed RAG service with connectors for Google Drive, Notion, Confluence, and more. Features include summary indexing, chunk reranking, hybrid semantic-keyword search, and multimodal indexing. Developers interact via simple SDKs; all embedding, storage, and retrieval infrastructure is abstracted.
Parameters—
Domainrag-retrievalagent-memory
Open SourceNo
WebsiteVisit
managed-ragconnectorsrerankingmultimodaldeveloper-API

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Long-Context Retrieval
1/5
RULER
no data
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
1/2
LoCoMo
57.89p
MemoryBank
no data
Cross-Session Memory
1/1
LongMemEval
57.76p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
61.59p
HotpotQA
63.927p
Agent Task Memory
1/1
AgentBench-Mem
59.53p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:Ragie vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Ragie vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Ragie vendor documentation; evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)Ragie vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Ragie vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Ragie vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)