AsterVec
What it is
AsterVec is an on-device vector database designed for AI agent memory and local RAG. It uses a disk-based HNSW graph and a separate LSM-tree for vector storage, allowing for high-speed search with a low, constant memory footprint.
The problem it solves
Standard in-memory vector databases (like HNSW) scale memory usage linearly with data size, making them impractical for large-scale local RAG or memory-constrained environments.
What you could use it for
- On-device AI agent memory
- Local RAG (Retrieval-Augmented Generation) applications
- Large-scale vector search with limited RAM
- Real-time vector updates and deletions without index rebuilding
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.