Jun 13Vibe with Hermes Agent — Bengaluru · RSVP
ToolsMCPBlogResearchCommunityStar on GitHub
All Tools
N
Vector DBFreeOpen Source

NEURONDB

AI-native vector search and ML inside PostgreSQL

ABOUT

Adding AI capabilities like vector search, embeddings, and ML inference to an application typically means running a separate vector database alongside your relational database — doubling infrastructure, adding ETL pipelines, and creating data consistency headaches. NeuronDB eliminates this by bringing vector similarity search (HNSW, IVFFlat), embedding generation, and ML inference directly into PostgreSQL as a native extension. Your vectors stay on your live rows, you query with standard SQL, and you get the same backups, replication, and security you already rely on. It also supports hybrid full-text + vector search for production RAG pipelines.

INTEGRATION GUIDE

1. Run semantic vector search on millions of rows without leaving PostgreSQL or managing a second database 2. Build a production RAG pipeline with hybrid full-text + vector retrieval using a single SQL query 3. Generate and store embeddings for text columns directly inside PostgreSQL with in-database inference 4. Power recommendation engines with kNN similarity queries over user behavior vectors 5. Run ML model training and prediction on live database rows without exporting data to a separate ML system

TAGS

vector-databasepostgresqlextensionembeddingssimilarity-searchmlhnswivfflat