Back to Arena

Activeloop Deep Lake

by Activeloop Inc.

System Card

OrganizationActiveloop Inc.
Released2021-02
Architecturevector-rag / multimodal AI data lakehouse
DetailsDeep Lake (by Activeloop) is an AI data runtime providing serverless multimodal data storage (tensors for text, images, audio, video, embeddings) alongside vector search. It is built on the Lance-compatible open format and integrates with PyTorch and TensorFlow for training data streaming as well as RAG retrieval. Supports both local and cloud-hosted deployments.
Parameters—
Domainrag-retrievallong-context
Open SourceYes
WebsiteVisit
multimodaldata-laketraining-datavector-searchPyTorch

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Long-Context Retrieval
4/5
RULER
70.746p
NIAH
no data
LooGLE
61.35p
LongBench
603p
∞Bench
69.33p
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
62.66p
MultiHop-RAG
no data
HotpotQA
60.615p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:Activeloop Deep Lake vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Activeloop Deep Lake vendor documentation; evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)Activeloop Deep Lake vendor documentation; evaluated on BABILong: Testing the Limits of LLMs with Long-Context Reasoning-in-a-Haystack (AIRI, 2406)Activeloop Deep Lake vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Activeloop Deep Lake vendor documentation; evaluated on InfiniteBench: Extending Long Context Evaluation Beyond 100K Tokens (Tsinghua / OpenBMB, 2402)Activeloop Deep Lake vendor documentation; evaluated on LooGLE: Can Long-Context Language Models Understand Long Contexts? (Peking University, 2311)