AI Factory Models

AI Factory Models is a core part of EDB Postgres® AI (EDB PG AI), giving you control over the entire lifecycle of your AI models:

  • Manage and govern model assets in your Asset Library
  • Deploy and scale inference services with Model Serving
  • Power Gen AI assistants, Pipelines, and RAG use cases
  • Enable Sovereign AI — using your models, your data, your infrastructure

Why Manage Models in EDB PG AI?

AI model management and serving is the foundation of trusted, governed AI. EDB PG AI allows you to:

  • Run inference in your clusters or controlled environments (Hybrid Manager-optimized)
  • Use models in Pipelines, Knowledge Bases, and Gen AI applications
  • Integrate private, commercial, or open source models
  • Maintain auditability and control over model usage

This enables enterprises to build Sovereign AI systems where models, data, and usage stay within trusted boundaries.

What You Can Do

  • Manage model images and metadata in Asset Library
  • Deploy models for inference via KServe and Model Serving
  • Track, update, and control model-serving infrastructure
  • Integrate models into AI Factory Pipelines and Gen AI Builder
  • Build hybrid architectures using private + external models

When to Use AI Factory Models

  • You want to run private model inference in your own clusters
  • You want full governance of models powering Pipelines and Knowledge Bases
  • You need trusted RAG workflows where embeddings match known models
  • You want to integrate specialized models into Gen AI apps and search
  • You want to provide internal APIs for model inferencing

Learn More

Manage Models

Deploy Models

EDB PG AI Models give you a trusted, governed foundation for all your AI-powered applications — with full transparency and operational control.


Deployment

Understand how model deployment works in AI Factory and how to deploy your models as scalable inference services.

Model Library

Manage and govern AI model images within the Hybrid Manager Asset Library for use in AI Accelerator Pipelines and Model Serving.

Model Serving

Deploy, manage, and scale AI model inference services in your EDB PG AI environment, powered by Hybrid Manager.


Could this page be better? Report a problem or suggest an addition!