Back to Arena

txtai

by NeuML

System Card

OrganizationNeuML
Released2020-08
Architecturevector-rag / Embedded vector database
DetailsAll-in-one embeddings database that bundles vector search, graph traversal, and LLM workflows in a single Python library. Backs many self-hosted RAG and memory deployments.
Parameters—
Domainrag-retrieval
Open SourceYes
WebsiteVisit
CodeRepository
embeddingslibrarypythonself-hosted

Capability Profile

Benchmark Scores

5 of 14 benchmarks
Long-Context Retrieval
2/5
RULER
75.786p
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
65.425p
HotpotQA
49.52p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
68.946p
Sources:txtai docstxtai benchmarktxtai (neuml/txtai); evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)txtai (neuml/txtai); evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)txtai (neuml/txtai); evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)