- Home
- Feature Store Software
- Agent Platform Feature Store
Agent Platform Feature Store
What it is
Agent Platform Feature Store is a paid, cloud-based feature store available through API and web interfaces. It manages machine-learning features and serves current values online from BigQuery data sources. BigQuery tables and views act as the offline store, while online serving supports predictions at low latency. The service offers online and offline stores, point-in-time joins, feature monitoring, and both serving modes. Data sources and feature columns may be registered in the Feature Registry to select columns, combine sources, and monitor statistics or drift. Monitoring can run manually or on a schedule to retrieve statistics and identify anomalies. Bigtable serving handles large volumes and near-real-time continuous synchronization, but does not support embeddings. Optimized online serving supports scheduled synchronization only and is scheduled to sunset on February 17, 2027. Google recommends Vector Search for embeddings. IAM policies govern access to feature groups, online store instances, and feature views. Resources must share a region or multi-region with their BigQuery source.
For and against
For
- BigQuery tables provide the offline store
- Low-latency online prediction serving
- Point-in-time joins are supported
- Scheduled or manual feature monitoring
- IAM access controls feature resources
Against
- No free plan
- Optimized serving is scheduled to sunset
- Bigtable serving does not support embeddings
- Resources must match BigQuery location
The verdict
Agent Platform Feature Store combines BigQuery-based offline data with managed online serving and monitoring. Check the serving-mode sunset and location requirements before choosing it.
Who it suits
It suits machine-learning teams that need managed feature storage, online prediction serving, BigQuery integration, and monitoring. It is also relevant for organizations controlling feature access with IAM policies.
Plans and prices
New Agent Platform Feature Store (pay-as-you-go)
Data processing node: $0.08 / 1 hour · Optimized online serving node: $0.30 / 1 hour · Bigtable online serving node: $0.94 / 1 hour · Bigtable online serving storage: $0.000342466 / 1 gibibyte hour
Read 4 Oct 2026 · cloud.google.com
On the record
- Data retention
- The online store has no data retention limit, while BigQuery retention limits or quotas may apply to the offline data.docs.cloud.google.com · read 4 Oct 2026
- Data sync limit
- Optimized online serving supports scheduled data sync but not continuous data sync; Bigtable online serving supports near-real-time continuous sync from BigQuery.docs.cloud.google.com · read 4 Oct 2026
- Embeddings
- Google recommends its purpose-built Vector Search to efficiently store and serve embeddings.docs.cloud.google.com · read 4 Oct 2026
- Feature registry
- Registering data sources and feature columns in the Feature Registry is optional; registration supports selecting columns, combining sources, and monitoring statistics and drift.docs.cloud.google.com · read 4 Oct 2026
- IAM access control
- IAM policies can control access to feature groups, online store instances, and feature views, with bindings for users, groups, domains, and service accounts.docs.cloud.google.com · read 4 Oct 2026
- Location constraint
- Feature Store resources must be in the same region or multi-region as the BigQuery source, and dual-region bucket sources are unsupported.docs.cloud.google.com · read 4 Oct 2026
- Monitoring
- Feature monitoring can run on a schedule or manually to retrieve statistics and detect anomalies such as feature drift.docs.cloud.google.com · read 4 Oct 2026
- Offline data
- BigQuery tables and views form the offline store, so Feature Store does not provision a separate offline store.docs.cloud.google.com · read 4 Oct 2026
- Online predictions
- It serves the latest feature values from BigQuery for online predictions at low latencies.docs.cloud.google.com · read 4 Oct 2026
- Purpose
- It is a managed, cloud-native feature store for managing ML features and serving them online from BigQuery data sources.docs.cloud.google.com · read 4 Oct 2026
- Serving options
- Bigtable online serving supports large data volumes but not embeddings; optimized online serving is deprecated and scheduled to sunset on February 17, 2027.docs.cloud.google.com · read 4 Oct 2026
- Support and learning
- The documentation links to Google Cloud Support, community forums, release notes, and tutorials with Colab and GitHub examples.docs.cloud.google.com · read 4 Oct 2026
