BigQuery now supports managed and SQL-native inference for Hugging Face and Vertex AI Model Garden open models.

Google Cloud Platform provides a suite of cloud computing services for building, deploying, and managing applications and infrastructure on Google's global network. Developers can learn about cloud-native development, machine learning, and big data analytics to leverage GCP's scalable and reliable cloud infrastructure for their projects.

Google Cloud

BigQuery now supports managed inference for open models from Hugging Face and Vertex AI Model Garden using SQL statements. Users can deploy models with a single CREATE MODEL command, run batch inference with AI.GENERATE_TEXT or AI.GENERATE_EMBEDDING functions, and manage resources automatically. The platform handles compute provisioning, automatically releases idle resources to control costs, and allows customization of machine types and replicas. The entire workflow from deployment to cleanup is managed through SQL, eliminating operational complexity for data teams working with LLMs and embedding models.

Introducing BigQuery managed and SQL-native inference for open models

A SQL-native workflow with automated management