BigQuery is your AI-ready data analytics platform

Connect your enterprise data to AI with a unified data analytics platform. BigQuery is designed to be multi-engine, multi-format, and multicloud, making it easier to store, analyze, and transform all your business data.

A single, unified data-to-AI platform

BigQuery is a unified data analytics platform that supports the end-to-end data life cycle. With BigQuery’s first-party integration with Vertex AI, you can tune, train, and ground multi-modal LLMs with enterprise data, without copying or moving data.


Simplicity and scale to manage all data and workloads in a single platform

Simplify, reduce cost, and risk of data workloads that do not work together. BigQuery has the simplicity and scale to manage structured, unstructured, and streaming workloads at the best price and performance.



Connect AI to more of your enterprise data

Bring gen AI to your data with scale and efficiency to leverage your business data with LLMs. BigQuery has first-party integration with Vertex AI to ground AI in the truth of your enterprise data.



Always-on intelligence for all your data teams

Increase the usage of actionable data to improve productivity. Gemini in BigQuery allows you to converse with your data in natural language and helps with code assist, recommendations, data exploration, and more.

Data analytics and AI in a single, unified experience

CategoryCapabilitiesHighlights
Build a data analytics foundation for AI

Customers increasingly want to run multiple analytics and AI use cases on a single copy of their data. BigQuery allows you to process data as easily in Python as you do with SQL, with a serverless Spark available directly in BigQuery. A unified metastore provides runtime metadata and connectors for SQL, open source engines, and AI/ML.

  • Fully serverless with no clusters to spin up or manage

  • Single user environment for all workloads

  • No data copying between different tools

BigQuery gives you the flexibility to use existing open source formats. BigLake, BigQuery's storage engine, provides a unified interface for analytics and AI engines to query multiformat, multicloud, and multimodal data. BigQuery supports Iceberg, Delta, and Hudi along with all processing engines and full capabilities over all of these table formats.

  • Query Apache Iceberg, Delta, and Hudi formats with fine-grained access control

  • Performance acceleration for Apache Iceberg with continuous optimization

  • Fully managed experience on Iceberg with support for high throughput streaming

BigQuery Studio is a one-stop shop for all data practitioners. BigQuery Studio has a great SQL editor as well as Python Notebooks. This allows your data teams their choice of SQL, Python, Spark, or natural language. Data teams can maximize productivity by collaborating with the Gemini-powered chat and code assistant within BigQuery.

  • Single unified workspace with Gemini-powered chat and code assistant

  • Use SQL, Python, Spark, Javascript, and natural language for analytics

  • Centralized source control and revision history

BigQuery makes it easy for you to manage, discover, and govern data with data governance capabilities built-in to BigQuery. This includes data quality, lineage, and profiling as well as governance rules to manage policies on BigQuery resources.

  • Data profiling to build deeper understanding of your data

  • Build trust in data at-scale with automatic data quality

  • Understand movement of data with end-to-end lineage

Ingest, process, and analyze event streams in real time to make data more useful and accessible with BigQuery's real-time capabilities. BigQuery continuous queries provides a real-time processing layer to analyze and transform incoming events in BigQuery. Customers can use Apache Kafka for BigQuery to manage streaming data workloads without the need to worry about version upgrades, rebalancing, monitoring, and other operational headaches.

  • Perform unbounded serverless analytics over streams of incoming data using SQL

  • Unified batch and streaming processing and real-time ETL with Dataflow

  • Ingest hundreds of millions of events per second with Pub/Sub

  • Managed Apache Kafka without the extra maintenance

Data-to-AI integration

BigQuery ML lets you create, train, and execute machine learning models using familiar SQL. It integrates with your choice of models including Gemini 1.0 Pro through Vertex AI, which is designed for high input/output scale and better result quality for text summarization or sentiment analysis tasks. You can build data pipelines that blend structured data, unstructured data, and generative AI models to create a new class of analytical applications.

  • Data-to-AI integration with inference engine and Vertex AI Model Registry

  • Modeling capabilities with ARIMA+ time series modeling, explainable AI, and more

  • Remote inference for LLMs to generate text and text embeddings

BigLake unifies data lakes and warehouses under a single management framework, enabling you to analyze, search, secure, govern and share unstructured data. Customers are already analyzing images using a broad range of AI models. BigLake has expanded capabilities to help you easily extract insights from documents and audio files using Vertex AI’s document processing and speech-to-text APIs.

  • Extract insights from documents and audio files

  • Create new apps for content generation, classification, or sentiment analysis

  • Create new apps for summarization, embeddings generation, and more

BigQuery vector search is integrated with Vertex AI to enable vector similarity search on your BigQuery data. This functionality can enable use cases like semantic search, similarity detection, and retrieval-augmented generation (RAG) with a LLM. Vector search can enhance the quality of your AI models by improving context understanding, reducing ambiguity, ensuring factual accuracy, and allowing adaptability to different tasks and domains.

  • Collect and standardize on vectors across multiple databases and Cloud Storage

  • Retrieve data relevant to a question and provide it with context to an LLM

  • Fully managed index keeps embeddings auto generated and in sync with Vertex

Enteprise capabilities

BigQuery automatically keeps a synchronous copy of your data in a second zone along with enough standby compute capacity to provide high availability in case of a data center level disaster. Cross-region disaster recovery provides managed failover in the unlikely event of a regional disaster. Cross-region disaster recovery will enable you to specify a reservation and a collection of datasets that BigQuery will maintain in a second region.

  • Guaranteed standby

  • Regional outage SLA

  • Coordinated failover

BigQuery helps you collaborate and securely exchange data assets at scale. You can create and manage environments for privacy-centric data sharing and analysis with data clean rooms. Data providers can manage subscriptions to data listings and monitor subscriber usage of shared data. You can share data across clouds with BigQuery Omni, and there is support for user defined functions, time travel, and materialized views over linked datasets.

  • Build a data clean room in a matter of clicks

  • Save costs and efficiently share data without the need to move it

  • Monitor usage of shared datasets

BigQuery Migration Service is a set of free tools to help you migrate to BigQuery. We continue to add new capabilities and now support different sources including Amazon Redshift, Apache HiveQL, Netezza, Teradata, Azure Synapse, Oracle, Presto, Snowflake, SQL Server, and Vertica. Generative AI-enhanced translations optionally aid the query compiler and automatically suggest output options with support for migrations from on-prem and cloud sources.

  • 15 query translation sources from popular data warehouses and data lakes

  • Four automated assessment sources with cost of ownership and migration effort

  • Generative AI-enhanced query translations

Data analytics and AI in a single, unified experience

Customers increasingly want to run multiple analytics and AI use cases on a single copy of their data. BigQuery allows you to process data as easily in Python as you do with SQL, with a serverless Spark available directly in BigQuery. A unified metastore provides runtime metadata and connectors for SQL, open source engines, and AI/ML.

  • Fully serverless with no clusters to spin up or manage

  • Single user environment for all workloads

  • No data copying between different tools

BigQuery ML lets you create, train, and execute machine learning models using familiar SQL. It integrates with your choice of models including Gemini 1.0 Pro through Vertex AI, which is designed for high input/output scale and better result quality for text summarization or sentiment analysis tasks. You can build data pipelines that blend structured data, unstructured data, and generative AI models to create a new class of analytical applications.

  • Data-to-AI integration with inference engine and Vertex AI Model Registry

  • Modeling capabilities with ARIMA+ time series modeling, explainable AI, and more

  • Remote inference for LLMs to generate text and text embeddings

BigQuery automatically keeps a synchronous copy of your data in a second zone along with enough standby compute capacity to provide high availability in case of a data center level disaster. Cross-region disaster recovery provides managed failover in the unlikely event of a regional disaster. Cross-region disaster recovery will enable you to specify a reservation and a collection of datasets that BigQuery will maintain in a second region.

  • Guaranteed standby

  • Regional outage SLA

  • Coordinated failover

Analyst reports

Gartner names Google Cloud a Leader

Gartner® recognizes Google Cloud as a Leader and positioned furthest in vision in the 2023 Magic Quadrant™ for Cloud Database Management Systems (DBMS).

Google Cloud is a Leader in the 2023 Forrester Wave

Google Cloud has been named a leader in The Forrester Wave™: Streaming Data Platforms, Q4 2023 report.

BigQuery gave us a solid data foundation for AI. Our data was exactly where we needed it. We were able to connect millions of customer data points from hotel information, marketing content, and customer service chat and use our business data to ground LLMs.

Allie Surina Dixon, Director of Data, Priceline

Take the next step

New customers get $300 in free credits.

Google Cloud
  • ‪English‬
  • ‪Deutsch‬
  • ‪Español‬
  • ‪Español (Latinoamérica)‬
  • ‪Français‬
  • ‪Indonesia‬
  • ‪Italiano‬
  • ‪Português (Brasil)‬
  • ‪简体中文‬
  • ‪繁體中文‬
  • ‪日本語‬
  • ‪한국어‬
Console
Google Cloud
  翻译: