Is there a free tier for BigQuery?

Yes. You get 1 TiB of query processing and 10 GB of storage free per month. New Google Cloud customers also receive $300 in initial credits.

How does BigQuery handle very large datasets?

It uses a massively parallel processing architecture with decoupled storage and compute, automatically scaling resources to handle petabyte-scale queries.

Can I use BigQuery for machine learning without exporting my data?

Yes. BigQuery ML allows you to create and execute ML models using standard SQL queries directly inside BigQuery, avoiding data movement.

How is data processed and what determines query cost?

Data is stored in a columnar format. On-demand pricing charges based on the total bytes processed in the columns selected by your query, calculated from data types.

What are 'slots' and how do they affect performance?

Slots are virtual CPUs used for query processing. In capacity mode, more slots mean higher concurrency and faster query execution. On-demand mode provides shared slots up to a limit.

Directory/Analytics/Data Warehousing/BigQuery

BigQuery

BigQuery is a fully managed, serverless, and highly scalable multi-cloud data warehouse from Google Cloud. It enables you to analyze petabytes of data using standard SQL or Python, with built-in machine learning, geospatial analysis, and business intelligence capabilities. The platform decouples storage and compute, allowing for independent scaling and cost optimization. Key capabilities include real-time analytics, data sharing, BigQuery ML for in-database machine learning, and integration with AI agents for automated data workflows. It supports both on-demand and flat-rate pricing models, making it suitable for ad-hoc analysis and predictable, high-throughput workloads. Target users include data analysts, data scientists, data engineers, and developers who need to derive insights from large datasets without managing infrastructure. It's designed for organizations looking for a unified analytics and AI platform.

Analytics

Visit website

Added on

Sep 22, 2026

ServerlessPetabyte-scaleOn-demand pricingSlot-based pricingGoogleSQLBigQuery MLVector searchApache Iceberg

Product information

Everything worth knowing about BigQuery

The complete picture — from everyday features to the technical detail builders and IT folks dig for.

Core features

What it actually does

Serverless data warehouse

Fully managed, petabyte-scale analytics platform with no infrastructure to set up or manage, using GoogleSQL or Python.

Built-in predictive analytics and AI inferencing

Train, evaluate, and deploy ML models directly within BigQuery using SQL. Integrate with AI functions for text summarization, sentiment analysis, and vector search.

Agentic AI for data teams

Get AI-powered assistance for data preparation, error detection, pipeline building, and conversational analytics via natural language.

Flexible compute pricing models

Choose between on-demand pricing (per TiB of data processed) and capacity pricing (reserved compute slots).

Real-time analytics with streaming

Use continuous queries, Kafka, and Dataflow for real-time data ingestion and analysis, including support for Apache Iceberg.

Technical capabilities

Decoupled storage-compute architecture

Leverages Google infrastructure like Dremel for fast SQL over massive datasets, with autoscaling slots and compressed columnar storage.

Integration with Google Cloud ecosystem

Native connectors to Cloud Storage, Pub/Sub, Datastream, Dataplex, and the broader AI/ML stack like Vertex AI.

Must watch videos

9:17

Tutorial9:17

Practical AI 010a: Using BigQuery ML

This is a demonstration of how to create a model using BigQuery ML. After testing the model, we look at the results and discuss some of the reasons the predictions aren't more accurate.

1:41

Tutorial1:41

BigQuery ML in a minute

Having fun watching these videos? Find more episodes by searching #CloudBytes ! Check out BigQuery ML → http://goo.gle/362gUH9 BigQuery ML lets you create and execute machine learning models in BigQuery using standard SQL queries - while democratizing machine learning and promoting application development. Watch to learn how you can use BigQuery ML for your machine learning projects! Watch more episodes of Cloud Bytes → https://goo.gle/3dFPPdM Subscribe to the GCP Channel → https://goo.gle/GCP Product: BigQuery ML, BQML; fullname: Stephanie Wong; #CloudBytes

How to simplify AI models with Vertex AI and BigQuery ML

3:39

Tutorial3:39

How to simplify AI models with Vertex AI and BigQuery ML

Vertex AI Workbench CodeLab → https://goo.gle/3Q38a8W Did you know there is native integration between Vertex AI and BigQuery ML? With unified cloud data, your machine learning pipelines will have multiple options for training and storing/accessing data. Watch along and learn about the new native integrations between Vertex AI and BigQuery ML for Google Cloud. Chapters: 0:00 - Intro 0:17 - What is BigQuery ML? 0:42 - What is Vertex AI? 1:26 - Vertex AI Model Registry 1:52 - Demo 2:46 - Wrap up 🏗️ Check out the playlist for more helpful AI hints → https://goo.gle/DeployingAI Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech #DeployingAI

Who uses it

Real use cases, no hype

Enterprise Data Warehousing

Centralize petabytes of structured and semi-structured data for unified, high-performance SQL analytics.

Real-time Analytics

Process and analyze streaming data from sources like Pub/Sub and Kafka for instant business insights.

Machine Learning & AI

Train, deploy, and serve ML models directly within the data warehouse using SQL (BigQuery ML) or integrated AI APIs.

Unstructured Data Analytics

Analyze images, PDFs, and video using built-in generative AI functions and object tables.

Data Migration

Migrate on-premises or cloud data warehouses (e.g., Teradata, Snowflake) with managed migration tools.

Embedded Analytics

Build and embed natural-language conversational analytics and BI dashboards into applications.

What's great

Truly serverless and fully managed—no infrastructure to manage.
Massive petabyte-scale analytics with automatic compute scaling.
Built-in ML/AI via BigQuery ML and Gemini integration.
Flexible pricing with on-demand (per TiB) and flat-rate (slots) models.

Technical strengths

Native support for querying unstructured data (images, PDFs) with AI functions.
Decoupled storage/compute architecture powered by Google's internal systems (Dremel, Colossus).
Seamless interoperability with open formats like Apache Iceberg.

Where it falls short

Complex pricing structure can lead to unexpected costs if not monitored.
Data ingestion and egress charges can add up quickly.
Limited SQL dialect (GoogleSQL) with some proprietary extensions.

Technical limitations

Cost can escalate significantly for frequent, large data scans.
On-demand model has concurrency limits (~2,000 slots) that can bottleneck heavy workloads.
Tightly coupled with GCP ecosystem, making multi-cloud strategies more complex.

For technical folks

The deep-dive specs

FAQ

Includes technical Q&A

On-demand charges per TiB of data scanned by your query. Capacity pricing charges for reserved compute slots (virtual CPUs) over time, offering more predictable costs for steady workloads.