HyperDB

Stale
GitHub Python MIT

Description

A hyper-fast local vector database for use with LLM Agents, providing lightweight vector storage and similarity search capabilities for embedding as instant memory and knowledge retrieval components in agent applications.

Key Features

  • Hyper-fast local vector database designed specifically for LLM Agents
  • C++ high-performance backend with hardware-accelerated operations via MKL BLAS
  • Simple unified interface compatible with all major LLM agent frameworks
  • Advanced document indexing with _id and _metadata support
  • Lightweight dependency, ready with pip install
  • Optional Hugging Face model local embedding support

Use Cases

💡 Agent instant memory storage and semantic retrieval
💡 Fast similarity search for local document knowledge bases
💡 Lightweight vector storage component in RAG applications
💡 Real-time conversation context fast recall

Quick Start

pip install hyperdb-python, import HyperDB, pass a document list to the constructor with a key field, call query() for similarity search, with save/load persistence support.

Related Projects