Product Launchllmembeddingsvertex aihuggingface
BigQuery Adds Managed Third-Party Model Inference
8.2
Google today announced managed third-party generative AI inference in BigQuery (Preview), enabling users to deploy and run Hugging Face and Vertex AI Model Garden models directly with SQL. The feature supports CREATE MODEL, AI.GENERATE_TEXT and AI.GENERATE_EMBEDDING, automated resource provisioning, idle-resource recycling via endpoint_idle_ttl, and manual lifecycle controls. It consolidates model lifecycle and cost management for analysts and ML engineers.
Key Points
- 1Enables running Hugging Face and Vertex AI Model Garden models directly from BigQuery using CREATE MODEL.
- 2Automates compute provisioning, idle-resource recycling and granular control to reduce costs and operational friction.
- 3Allows analysts and ML engineers to perform SQL-native inference and embedding generation at scale.
Scoring Rationale
Official Google feature launch with broad usability; limited novelty beyond convenience and existing model access.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
High-Value Overnight OrdersEasyDelivered International ShipmentsMediumOn-Time Delivery Rate by CarrierHard
250 free problems · No credit card
See all Logistics & Shipping problems