Feature stores · Amazon Web Services
Amazon SageMaker Feature Store
Fully managed feature repository inside SageMaker that stores, discovers and serves ML features for training and inference.
Amazon SageMaker Feature Store is the feature-management module of the SageMaker ML platform rather than a standalone product. Teams ingest features from S3, Redshift, Snowflake, Databricks Delta Lake, application logs or clickstreams into named feature groups, optionally applying a transformation at ingest time (counts, time-window aggregates). Feature groups keep both an online store for low-latency inference lookups and an offline store for point-in-time-correct training datasets, addressing the standard feature-store problem of keeping training and serving values consistent. Studio's visual interface makes feature groups searchable and taggable so teams can reuse existing features instead of rebuilding pipelines, and groups can be shared and discovered across AWS accounts. It only exists inside SageMaker and is billed as part of AWS usage rather than sold or priced separately.
At a glance
| Vendor | Amazon Web Services |
|---|---|
| Pricing model | Usage-based |
| Free tier | — |
| Deployment | Cloud |
| Open source | No |
| Best for | Teams already building models on SageMaker who need a managed registry to reuse features across the ML lifecycle. |
Pricing
Not sold separately: billed as AWS usage for online/offline storage and read/write throughput within a SageMaker account.
Pricing has not been verified yet — see the vendor's site.
Features
- Feature groups with combined online and offline stores
- Point-in-time-correct training dataset retrieval
- Low-latency online lookups for real-time inference
- Ingest-time feature transformations (counts, windowed aggregates)
- Discovery and tagging of feature groups in SageMaker Studio
- Cross-account feature group sharing
- Ingestion from S3, Redshift, Snowflake and Delta Lake
Integrations
Profile last reviewed September 21, 2026