Back to Arena

DB-GPT

by eosphoros-ai

System Card

Organizationeosphoros-ai
Released2023-05
Architectureagentic-workflow / Data-centric agent with sandbox memory
DetailsAgentic data assistant connecting to DBs and files, writing SQL, running Python analysis in sandboxes, generating reports. Multi-LLM + multi-source connectors; skills extensibility; persistent conversation and data memories.
Parameters—
Domainagent-memoryknowledge-graph
Open SourceYes
WebsiteVisit
CodeRepository
data-agentsqlsandboxreports

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
1/2
LoCoMo
80.688p
MemoryBank
no data
Cross-Session Memory
1/1
LongMemEval
77.761p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
7579p
HotpotQA
72.657p
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
70.655p
Sources:DB-GPT (eosphoros-ai/DB-GPT); evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)DB-GPT (eosphoros-ai/DB-GPT); evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)DB-GPT (eosphoros-ai/DB-GPT); evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)DB-GPT (eosphoros-ai/DB-GPT); evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)DB-GPT (eosphoros-ai/DB-GPT); evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)DB-GPT (eosphoros-ai/DB-GPT); evaluated on RAGAS: Automated Evaluation of Retrieval-Augmented Generation (Exploding Gradients, 2309)