Model Registry

Version control for AI models.

Centralized registry for ML models. Version control, lineage tracking, approval workflows, and deployment automation.

Create registry View docs
v1.0.0Devacc: 0.94 | f1: 0.91v1.1.0Stagingacc: 0.96 | f1: 0.93v1.2.0Productionacc: 0.97 | f1: 0.95

Unlimited

Models

Immutable

Versioning

Full

Lineage

One-click

Deploy

Models, governed.

Version control with lineage and approvals.

Model versioning

Immutable versions with semantic version tags.

Lineage tracking

Track data, code, and params that produced each model.

Approval workflows

Stage gates for model promotion: Dev → Staging → Prod.

One-click deploy

Deploy any model version to inference endpoint.

Model cards

Auto-generated model cards with metrics and biases.

Format support

ONNX, TensorFlow, PyTorch, PMML, and more.

Getting started

Launch your first instance in three steps. CLI, console, or API — your choice.

Terminal
ur ml registry push fraud-model \
  --version=1.2.0 \
  --artifact=model.onnx

Registry patterns.

Governance and model deployment.

Model governance

Centralized model management with approval gates.

View tutorial

Suggested configuration

Versioning · Lineage · Approvals

Estimate your costs

Create detailed configurations to see exactly how much your architecture will cost. Pay for what you use, down to the second.

Configuration 1

Estimated: $90.00/mo

Model Registry

Usage

min/mo

Storage & Artifacts

GB
Security ScanningVulnerability scanning for artifacts
Config 1 cost$90.00

Cost details

$90.00

Version control, lineage, deployment automation.

Configuration 1
$90.00
10 Seat(s)$80.00
Usage$5.00
Storage$5.00
Share

Works seamlessly with

Training
MLOps
Inference
IAM
Monitoring
Analytics

Frequently asked questions

Models, governed.

Version control with lineage and approvals.

Create registry View docs