Gemini in BigQuery overview

This document describes how Gemini in BigQuery, which is part of the Gemini for Google Cloud product suite, provides AI-powered assistance to help you work with your data.

AI assistance with Gemini in BigQuery

Gemini in BigQuery provides AI assistance to help you do the following:

• Explore and understand your data with data insights. Generally available (GA) Data insights offers an automated, intuitive way to uncover patterns and perform statistical analysis by using insightful queries that are generated from the metadata of your tables. This feature is especially helpful in addressing the cold-start challenges of early data exploration. For more information, see Generate data insights in BigQuery.
• Discover, transform, query, and visualize data with BigQuery data canvas. (GA) Using natural language, you can find, join, and query table assets, visualize results, and seamlessly collaborate with others throughout the entire process. For more information, see Analyze with data canvas.
• Get assisted SQL and Python data analysis. You can use Gemini in BigQuery to generate or suggest code in SQL or Python, and to explain an existing SQL query. You can also use natural language queries to begin data analysis. To learn how to generate, complete, and summarize code, see the following documentation:
  • Use the SQL generation tool (GA)
  • Prompt to generate SQL queries (GA)
  • Complete a SQL query (Preview)
  • Explain a SQL query (GA)
  • Generate Python code (GA)
  • Python code completion (Preview)
• Prepare data for analysis. (Preview) Data preparation in BigQuery gives you context aware, AI-generated transformation recommendations to cleanse data for analysis. For more information, see Prepare data with Gemini.
• Optimize your data infrastructure with partitioning, clustering, and materialized view recommendations. You can let BigQuery monitor your SQL workloads for opportunities to improve performance and reduce costs. For more information, see the following documentation:
  • Manage partition and cluster recommendations (GA)
  • Manage materialized view recommendations (Preview)
• Autotune and troubleshoot serverless Apache Spark workloads. (Preview) Autotuning can automatically optimize Spark jobs by applying configuration settings to a recurring Spark workload based on best practices and an analysis of prior workload runs. Advanced troubleshooting with Gemini in BigQuery can explain and surface job errors, and it can offer actionable recommendations to fix slow or failed jobs. For more information, see Autotuning Spark workloads and Advanced troubleshooting.
• Customize your SQL translations with translation rules. (Preview) Create Gemini-enhanced translation rules to customize your SQL translations when using the interactive SQL translator. You can describe changes to the SQL translation output using natural language prompts or specify SQL patterns to find and replace. For more information, see Create a translation rule.

The Gemini large language models (LLMs) that are used by Gemini in BigQuery are trained on datasets of publicly available code, Google Cloud-specific material, and other relevant technical information in addition to the datasets used to train the Gemini foundation models. Models are trained so that Gemini in BigQuery responses are as useful to Gemini in BigQuery users as possible.

Learn how and when Gemini for Google Cloud uses your data. As an early-stage technology, Gemini for Google Cloud products can generate output that seems plausible but is factually incorrect. We recommend that you validate all output from Gemini for Google Cloud products before you use it. For more information, see Gemini for Google Cloud and responsible AI.

Note: Gemini in BigQuery is part of Gemini for Google Cloud and doesn't support the same compliance and security offerings as BigQuery. You should only set up Gemini in BigQuery for BigQuery projects that don't require compliance offerings that aren't supported by Gemini for Google Cloud. For information about how to turn off or prevent access to Gemini in BigQuery, see Turn off Gemini for Google Cloud products.

Pricing

See Gemini for Google Cloud pricing.

Quotas and limits

For quotas and limits that apply to Gemini in BigQuery, see Gemini for Google Cloud quotas and limits.

Where to interact with Gemini in BigQuery

After you set up Gemini in BigQuery, you can use Gemini in BigQuery to do the following in BigQuery Studio:

• To use data insights, go to the Insights tab for a table entry, where you can identify patterns, assess quality, and run statistical analysis across your BigQuery data.
• To use data canvas, create a data canvas or use data canvas from a table or query to explore data assets with natural language and share your canvases.
• To use natural language to generate SQL or Python code, or receive suggestions with autocomplete while typing, use the SQL generation tool for your SQL queries or Python code. Gemini in BigQuery can also explain your SQL code in natural language.
• To prepare data for analysis, in the Create new list, select Data preparation. For more information, see Open the data preparation editor in BigQuery.
• To view recommendations for partitioning, clustering, and materialized views, click Recommendations in the Google Cloud console toolbar.

Autotune and troubleshoot Spark jobs

Autotuning can help you optimize your Spark workloads for performance and resilience. Instead of manually configuring settings, Gemini in BigQuery can apply best practices for recurring workloads and then help you understand and monitor your autotuning. Advanced troubleshooting provides natural language answers to "What was autotuned?", "What is happening now?", and "What can I do about it?"

Set up Gemini in BigQuery

For detailed setup steps, see Set up Gemini in BigQuery.

What's next

• Learn how to set up Gemini in BigQuery.
• Learn how to write queries with Gemini assistance.
• Learn how Gemini for Google Cloud uses your data.
• Learn more about Google Cloud compliance.