BigHugger
GH Repository · activeloopai

deeplake

Deeplake is AI Data Runtime for Agents. It provides serverless postgres with a multimodal datalake, enabling scalable retrieval and training.

stars
9,235
30-day movement
+21/day
Related entries
60
Connections
1
agent-appC++ragskillmultimodalfilesystemopenclawdeep-learningaillmmlopscomputer-visionlarge-language-modelsmakevector-databasepostgresmemorydatalakeagentic-ragagentpytorchclawbot

Deeplake, by activeloopai, describes itself as an AI Data Runtime for Agents, offering serverless Postgres combined with a multimodal datalake. Its repository topics indicate support for scalable retrieval and training workflows, including vector database, RAG, and memory use cases.

You reach for it when your agent applications need a unified data runtime that combines Postgres-like querying with multimodal storage for retrieval and training.

Use it to

  • Store and retrieve multimodal data for agent memory
  • Build agentic RAG pipelines with a vector database
  • Serve data via a serverless Postgres interface
  • Prepare datasets for deep-learning training in PyTorch
  • Manage LLM-related datalake workloads

For ML and AI engineers building data-driven agent applications

Role
agent-app
Language
C++
Licence
Apache-2.0
Forks
722
Open issues
53
Last push
2026-05-21
Latest release
1.0.6 · 2020-12-15
topicsaidatalakevector-databasepostgresagentic-ragmemory