PostgresML

PostgresML is a database extension that integrates machine learning and AI capabilities directly into PostgreSQL, enabling users to index, filter, and rank vectors, create embeddings, and run inference and grounded answers in SQL. It targets developers and data teams at companies of all sizes, particularly those building AI applications. It is positioned as a solution that simplifies the AI stack by keeping vectors, models, and application data in one GPU-backed database, reducing complexity and improving performance compared to separate microservices.

Key features

• Index, filter and re-rank vector embeddings
• 10x faster vector operations
• Fast KNN and ANN search
• Index embeddings with HNSW or IVFFlat
• Generate embeddings from state-of-the-art models
• Built-in data preprocessors for splitting and chunking
• Colocate data and compute
• Embed, serve and store all in one process
• Terabytes of data on a single machine
• Built-in data privacy & security
• Train, tune and deploy regression, classification, clustering
• Fine-tune LLMs on your own data
• Monitor model deployment