Skip to main content

Model Serving

Simplified Production ML

illustration-nodes-1-gray
model-serving-intro

Introduction

Model Serving is built within the Databricks Lakehouse Platform and integrates with your lakehouse data, offering automatic lineage, governance and monitoring across data, features and model lifecycle. Simplify model deployment, reduce infrastructure overheads and accelerate time to production. With built-in auto-scaling capability, you can scale from zero all the way up to your most critical needs while paying only for the compute you use.

Customer Quotes

unified-with-lakehouse-data

Unified with lakehouse data

Deep integration with the lakehouse provides automatic lineage, governance and monitoring across data, features and model lifecycle

simplified-deployment

Simplified deployment

Automated infrastructure configuration and maintenance to reduce overheads and accelerate your ML deployments

real-time

Real-time

Quickly scale from zero all the way up to your most critical needs, and down as your needs change, while paying only for the compute you use

Get started with these resources

eBook

mlops-virtual-event-thumb

Ready to get started?