Use this page to understand the differences between Vertex AI andBigQuery and learn how you can integrateVertex AI with your existing BigQuery workflows.Vertex AI and BigQuery work together to meet your machinelearning and MLOps use cases.

To learn more about model training differences between Vertex AI andBigQuery,seeChoose a training method.

Differences between Vertex AI and BigQuery

This section covers the Vertex AI, BigQuery, andBigQuery ML services.

Vertex AI: An end-to-end AI/ML platform

Vertex AI is an AI/ML platform for model developmentand governance. Common use cases include the following:

• Machine learning tasks, such as forecasting, prediction, recommendation,and anomaly detection

• Generative AI tasks, such as:

  • Text generation, classification, summarization, and extraction
  • Code generation and completion
  • Image generation
  • Embedding generation

You can use BigQuery to prepare training data forVertex AI models, which you canmake available as features in Vertex AI Feature Store.

You can train models in Vertex AI in three ways:

• AutoML: Train models on image,tabular, and video datasets without writing code.
• Custom Training: Runcustom training code catered to your specific use case.
• Ray on Vertex AI: Use Ray to scale AI and Python applications like machine learning.

You can also import a model trained on another platform likeBigQuery ML or XGBoost.

You can register custom-trained models to theVertex AI Model Registry.You can also import models trained outside of Vertex AI and register themto Vertex AI Model Registry. You don't need to registerAutoML models; they are registered automatically at creationtime.

From the registry, you can manage modelversions, deploy to endpoints for online predictions, perform modelevaluations, monitor deployments with Vertex AI Model Monitoring, anduseVertex Explainable AI.

Available languages:

• TheVertex AI SDKsupports Python, Java, Node.js, and Go.

BigQuery: A serverless, multicloud enterprise data warehouse

BigQuery is a fully managed enterprisedata warehouse that helps you manage and analyze your data with built-in featureslike machine learning, geospatial analysis, and business intelligence.BigQuery tables can be queried by SQL, and data scientists who primarilyuse SQL can run large queries with only a few lines of code.

You can also use BigQuery as a data store that you reference whenbuilding tabular and custom models in Vertex AI. To learn more aboutusing BigQuery as a data store, seeOverview of BigQuerystorage.

Available languages:

• SDKs for BigQuery. To learn more, see theBigQuery API Client Libraries.
• GoogleSQL
• Legacy SQL

To learn more, seeBigQuery SQL dialects.

BigQuery ML: Machine learning directly in BigQuery

BigQuery ML lets you develop and invoke models inBigQuery. With BigQuery ML, you can use SQL totrain ML models directly in BigQuery without needing to movedata or worry about the underlying training infrastructure. You can createbatch predictions for BigQuery ML models to gain insights fromyour BigQuery data.

You can also access Vertex AI models by usingBigQuery ML. You can create a BigQuery MLremote model over aVertex AI built-in model like Gemini,or over aVertex AI custom model. You interact with the remote model usingSQL in BigQuery, just like any other BigQuery MLmodel, but all training and inference for the remote model is processed inVertex AI.

Available language:

• GoogleSQL
• BigQuery client libraries

To learn more about the advantages of using BigQuery ML, seeIntroduction to AI and ML in BigQuery.

Benefits of managing BigQuery ML models in Vertex AI

You can register your BigQuery ML models to theModel Registry in order to manage the models inVertex AI. Managing BigQuery ML models inVertex AI provides two main benefits:

• Online model serving: BigQuery ML only supports batch predictionsfor your models. To get online predictions, you can train your models inBigQuery ML and deploy them to Vertex AI endpoints throughVertex AI Model Registry.

• MLOps capabilities: Models are most beneficial when they are kept up todate through continuous training. Vertex AI offers MLOps tools thatautomate the monitoring and retraining of models to maintain the accuracyof predictions over time. With Vertex AI Pipelines, you can useBigQuery operators to plug any BigQuery jobs (includingBigQuery ML) into an ML pipeline. WithVertex AI Model Monitoring, you can monitor your BigQuery MLpredictions over time.

To learn how to register your BigQuery ML models to the Model Registry,seeManage BigQuery ML models with Vertex AI.

Related notebook tutorials

What's next

• To get started with Vertex AI see:
  • Choose a training method
  • Integrate a BigQuery ML model with Model Registry