Back to Arena

Dust tt

by Dust (formerly XP1)

System Card

OrganizationDust (formerly XP1)
Released2023-04
Architectureagentic-workflow / Horizontal Enterprise Agent Infrastructure
DetailsDust is a fully open-source TypeScript/Next.js platform that connects to company knowledge sources (Slack, Notion, Confluence, GitHub) via real-time Temporal Workflow ingestion pipelines and builds per-workspace RAG indexes. Teams can create no-code agents that share a unified permission-aware knowledge base, with each agent able to read, write, and act across connected tools. Deliberately chose horizontal infrastructure over vertical domain solutions.
Parameters—
Domainagent-memoryrag-retrieval
Open SourcePartial
WebsiteVisit
open-sourceenterprise-agentsno-codeconnectorstemporaltypescript

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
77.570p
MemoryBank
no data
Cross-Session Memory
1/1
LongMemEval
7545p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
72.759p
HotpotQA
72.557p
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:Dust tt vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Dust tt vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Dust tt vendor documentation; evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)Dust tt vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)Dust tt vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Dust tt vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)