Cloud Equivalents

Vision & documents · AWS vs Google Cloud

Amazon Rekognition Custom Labels vs Agent Platform AutoML

Rekognition Custom Labels maps to AutoML image classification and object detection on Agent Platform.

Direct match Use it to: Train your own model to recognize your products, defects, or logos from labeled images.

Last verified: · Edit on GitHub · Report an error

Side-by-side comparison

Custom image classification & detection: AWS vs Google Cloud side by side
AspectAmazon Web ServicesGoogle Cloud
ServiceAmazon Rekognition Custom LabelsAgent Platform AutoML
LifecycleGenerally availableAutoML Text shut down Jun 15, 2025 and AutoML Video shut down Jul 31, 2025. Tabular and image AutoML remain.
What it isTrain custom image classification and object detection models from labeled images.Trains tabular (classification, regression, forecasting) and image models from your data with little or no code.
Pricing modelBilled per training hour and per inference hour while the model is running.Training and serving are billed per node-hour.
DocsAmazon Rekognition Custom Labels docsAgent Platform AutoML docs
Pricing pageAmazon Rekognition Custom Labels pricingAgent Platform AutoML pricing

Key differences

  • Trained models are deployed to endpoints (node-hours) or exported, instead of being started and stopped like Custom Labels models.

Pricing summary

Pricing models only. No prices are quoted because they change by region and over time.

AWS
Billed per training hour and per inference hour while the model is running. Amazon Rekognition Custom Labels pricing →
GCP
Training and serving are billed per node-hour. Agent Platform AutoML pricing →

And on Azure?

Partial match Azure's equivalent is Azure Custom Vision, Azure ML automated ML (AutoML). AWS vs Azure comparison →

FAQ

What is the Google Cloud equivalent of Amazon Rekognition Custom Labels?

Rekognition Custom Labels maps to AutoML image classification and object detection on Agent Platform.

How does pricing compare between Amazon Rekognition Custom Labels and Agent Platform AutoML?

AWS: Billed per training hour and per inference hour while the model is running. Google Cloud: Training and serving are billed per node-hour. Check the official pricing pages for current rates.

What is the Azure equivalent?

Azure Custom Vision and Azure ML automated ML (AutoML) (partial match). Rekognition Custom Labels maps to Azure Custom Vision, but Custom Vision is retiring (supported until September 25, 2028).

Related AWS vs GCP comparisons

Official sources