Tibor Kiss

Today we’re unveiling new AI capabilities to help cloud developers and operators at every step of the application lifecycle. We are doing this by:

• Putting applications at the center of your cloud experience, abstracting away the infrastructure complexities of the traditional cloud model. Now you can design, observe, secure, and optimize at the application level, rather than at the infrastructure level.

• Adding AI assistance everywhere in the application development lifecycle with new features in Gemini Code Assist and Gemini Cloud Assist. We’re introducing new agents, including an agent designed for complex workflows, to offload development tasks in your IDE. We’re also applying AI to management and operational tasks to make your entire application journey smarter and more productive.

The new application-centric cloud experience 

Applications are the engines of modern businesses. However, the traditional resource-centric cloud model introduces a huge amount of complexity to application design, deployment, and management, forcing developers to translate business requirements into lower-level infrastructure details, obscuring the application’s overall purpose and performance goals. Application components are spread across many systems and dashboards, making it hard to track performance, understand costs, and troubleshoot issues. Our application-centric approach helps you address these challenges, with a number of new services and expanded features.

Introducing Application Design Center

Our new service, Application Design Center, now in public preview, helps platform admins and developers streamline the design, deployment, and evolution of cloud applications, while helping ensure that they are secure, reliable, and aligned with best practices. In addition to APIs and the gcloud command line interface, Application Design Center provides a visual, canvas-style approach to designing and modifying application templates. It also lets you configure application templates for deployment, view infrastructure as code in-line, and collaborate with teammates on designs. Application deployments are automatically registered in App Hub for easy operations and troubleshooting.

Cloud Hub unifies visibility and control of your applications

The new Cloud Hub service is the central command center for your entire application landscape, allowing you to focus on what matters the most for your applications and workloads. It helps you manage the key aspects of your applications, with insights into deployments, health and troubleshooting, resource optimization, maintenance, quotas and reservations, and support cases. This unified view simplifies management and accelerates problem resolution. Cloud Hub is currently in public preview and accessible to all Google Cloud console users. Experience Cloud Hub here.

Expanding App Hub and app-centric observability

App Hub models your applications as interconnected services and workloads for an application-centric view, and is now integrated with over 20 Google Cloud products, including Google Kubernetes Engine (GKE), Cloud Run, Cloud SQL, and AlloyDB, strengthening the foundation of our application-centric experience. And we will continue to expand from here. 

Additionally, we are enhancing our application-centric observability capabilities with two new features: first, Application Monitoring, in public preview, which supports automatically tagging telemetry (logs, metrics, and traces) with application context, application-aware alerts, and out-of-the-box application dashboards; second, we’re introducing Cost Explorer to provide visibility into granular application costs and utilization metrics, allowing you to identify efficiency opportunities. Cost Explorer is currently in private preview; sign up here to try it out.

AI-assisted application development and management 

With Gemini Code Assist and Cloud Assist, we are focused on helping you accelerate application development and streamline your cloud operations across your entire software development lifecycle.

Gemini Code Assist: boost developer productivity

Gemini Code Assist already provides AI assistance in your favorite code editors – Android Studio, Firebase Studio, JetBrains, VS Code, and others. Today we’re announcing three new ways that Code Assist can help you as a software engineer:

• Agents that can translate natural language requests into multi-step, multi-file solutions

• New tools that make it easy to connect Code Assist to external services, third-party partners, or even other agents

• Gemini 2.5, with its enhanced coding capabilities, now available in Gemini Code Assist for individuals

Gemini Code Assist agents

Not long ago, Sundar Pichai shared that 25% of Google’s code is written by AI. Our customers are on a similar trajectory. Ampere from Renault Group, Broadcom, CGI, CME Group, PayPal, and Wipro use Code Assist today for writing code and developing applications. In fact, CME Group, which runs the Chicago Mercantile Exchange, states that Code Assist provides most of its developers more than 10.5 hours of productivity gain per month.

Getting an application to production involves a whole range of tasks beyond coding. That’s why we’re putting powerful new capabilities at your fingertips, with Gemini Code Assist agents designed to help you to tackle tasks across the entire software development lifecycle.

Gemini Code Assist’s agents can now help you do all kinds of common developer tasks:

1. Generate new software: Create applications from product specifications in Google Docs.

2. Migrate code: Perform code transformation from one language to another and code translation between language/framework versions.

3. Implement new features: Read issues assigned to users in GitHub and create code to address them. 

4. Execute code reviews: Provide code analysis and feedback on quality, best practices, and organizational style.

5. Generate tests: Create test cases, execute tests, and report results.

6. Perform AI testing: Conduct an evaluation of model inputs/outputs to identify and address potential content safety concerns.

7. Generate documentation: Write and update wiki-style documentation for entire repositories and provide full codebase Q/A.

When agents are doing work in the background, we know it’s important to be able to understand which tasks are underway and what they’ve completed. That’s why we created the Gemini Code Assist Kanban board, which lets you interact with our agents, review the workplan that Gemini creates to complete the tasks, and track the progress of the various jobs/requests.

Gemini Code Assist tools

To make your work in the IDE with Gemini Code Assist truly effective, you need easy access to context from outside sources. That’s why we built Gemini Code Assist tools, now in preview. These are prebuilt connections accessible within Gemini Code Assist’s chat that help you access information from Google apps and industry-leading tools from partners including Atlassian, Sentry, Snyk, and more.

More IDEs: Android and Firebase Studio

Gemini Code Assist helps you write, transform, and understand code directly in the IDE of your choice, and we’re excited to announce that this coverage is expanding. Gemini Code Assist is now available in Android Studio for professional developers who want AI coding assistance with enterprise security and privacy features.

In addition, we launched two new purpose-built agents for the Firebase platform powered by Gemini Code Assist: an App Prototyping agent within the new Firebase Studio development environment that turns your app ideas into fully functional prototypes, including the UI, backend code, and AI flows; and an App Testing agent within Firebase App Distribution that prepares mobile apps for production by generating, managing, and executing end-to-end tests. Both agents are currently in preview.

Gemini Cloud Assist: simplify application operations and management

Gemini Cloud Assist provides AI-powered assistance across the application lifecycle in the Google Cloud environment, helping developers, DevOps teams, cloud admins, and SREs manage applications more effectively and efficiently — from initial infrastructure design, deployment, monitoring, and troubleshooting, to continuous optimization. Gemini Cloud Assist offers AI assistance through a chat interface, within purpose-built AI-guided workflows, and directly inside Google Cloud products. It is available both in the console, and via mobile. Gemini Cloud Assist is currently in public preview and accessible for every Google Cloud user to try.

Design applications

Gemini Cloud Assist is integrated with Application Design Center to accelerate application infrastructure design and deployment. Using Gemini Cloud Assist, you can design your application infrastructure by describing your desired outcomes in natural language. For example, you can ask Gemini Cloud Assist to “design a three-tier e-commerce website,” or “design a mobile app which converts photos to videos,” and it automatically generates architecture diagrams and application templates, which you can iteratively refine with Gemini. With the deployment workflow directly built into the Application Design Center, you can also instantly deploy these application templates onto Google Cloud. In addition, you can distribute these templates to your team using privately shared catalogs, or export them as Terraform for broader interoperability with other workflows.

Troubleshoot with Investigations

Gemini Cloud Assist Investigations is a new AI-powered workflow that significantly accelerates troubleshooting and issue resolution. It leverages your cloud environment and a comprehensive set of data such as log and error patterns, configuration changes, and troubleshooting runbooks, to quickly identify root causes and provide mitigation options. 

The Investigations flow includes errors seen, affected resources, and user-provided symptoms that lead to observations and summarized findings generated by Gemini Cloud Assist. You can launch Investigations directly from errors and log entries, or via the Investigations icon at the top-right corner of every console page. Then, you can ask questions in chat related to a given investigation and iterate through revisions. In addition, Gemini Cloud Assist facilitates an easy handoff of Investigations cases to Google Support, who, with your consent, can leverage the findings to provide continued human assistance. Gemini Cloud Assist Investigations is currently in private preview. To try it, sign up here.

Optimize cost and utilization

Gemini Cloud Assist also provides AI-driven cost management and optimization capabilities for developer and DevOps teams, as well as for FinOps practitioners and finance teams. In the Cloud Hub Cost Optimization dashboard, developers and DevOps teams get insights into application cost and utilization. By linking costs with utilization data like vCPU or memory, Gemini identifies inefficiencies and delivers personalized recommendations. For more granular cost and utilization analysis, Cost Explorer provides flexible heatmap visualizations and aggregates compute and memory utilization across single or multiple resources. It supports 30 days of historical data and is fully filterable by project, application, Google Cloud product, and resource type.

Within Cloud Billing, we have already seen over 100,000 FinOps hours saved by our customers using Gemini Cloud Assist cost reports over the past year (Jan 24-Jan 25). We are now extending Gemini Cloud Assist to include waste insights and cost optimization opportunities with FinOps Hub 2.0.

Gemini Cloud Assist integrations across Google Cloud 

We’re providing AI-powered assistance directly within your existing workflows by integrating Gemini Cloud Assist across your favorite Google Cloud services, including Storage Insights, Cloud Observability, Firebase, Database Center, Flow Analyzer, FinOps Hub, as well as our security- and compliance-related services.

This also includes an improved chat experience. Gemini Cloud Assist is now able to provide even better quality assistance with richer context about your deployments, including your assets and their configurations, system health, metrics and logs, and cost reporting, as well as addressing specialized domain-specific tasks such as BigQuery code generation, and database and storage explanation.

Build and manage your applications with Google Cloud today

Our new application-centric cloud powered by AI makes it easier to build and manage the applications that fuel the future of your organization.

To help get you going you can try out the new Enterprise tier of the Google Developer Program in limited preview. This new offer provides developer and platform teams a safe and affordable way to explore Google Cloud and its AI products. For a set monthly cost of $75/month per seat, developers get access to a sandbox environment where they can experiment with new technologies, build proof of concepts, and improve their skills. It includes Gemini Code Assist Enterprise, discounts on cloud resources, and integrations with Gitlab, Harness, Jetbrains, and Snyk, so developers can take advantage of reference architectures to build common gen AI application patterns. You can learn more about this offer here.

And to try the new products and features announced in this blog: 

• Log into your Google Cloud console today to experience the power of Gemini Cloud Assist. Press “Opt(Alt)+G” or click the Gemini button to launch Gemini Cloud Assist in the console. To learn more details about the features, visit the website here.

• Log into Cloud Hub in Google Cloud console today to get full visibility of your application landscape. To try the Cost Optimization dashboard of Cloud Hub, sign up here.

• Sign up for the Gemini Code Assist agents preview waitlist.

• Sign up to try Gemini Cloud Assist Investigations.

• Install the Gemini Code Assist for individuals extension in your IDE of choice to use Gemini 2.5 for your personal projects.

• Check out Firebase Studio.

The AI era is here, fundamentally reshaping industries and demanding unprecedented network capabilities for training, inference and serving AI models. To power this transformation, organizations need global networking solutions that can handle massive capacity, seamless connectivity, and provide robust security. 

At Next 25, we’re addressing these critical needs and enabling customers to build and deliver distributed AI applications with ease through a suite of innovations in our cloud networking products and Cross-Cloud Network solutions.

These innovations include AI-optimized networking, simplified and secure service networking, and zero-trust security against zero-day threats. We are also expanding our Cross-Cloud Network solutions to include programmability and performance for the Global Front End for web, media, and generative AI services, plus our newest solution, Cloud WAN, which provides a fully managed global network for secure, simplified connectivity across enterprise locations, powered by our extensive global infrastructure.

AI-optimized networking: performant, secure, scalable

To make your AI models work their best, you need a network that can handle massive amounts of data and intense processing. Whether you’re training huge models or serving them to users (“inference”), speed, reliability, and security are essential. You’re dealing with complex infrastructure, and moving tons of data to provide lightning-fast responses. Our innovations are focused on enabling the infrastructure you need for these demanding AI workloads:

• Massive data ingestion with 400G Cloud Interconnect and Cross-Cloud Interconnect: Onboard your AI datasets faster and train cross-cloud with 4x the bandwidth of our 100G Cloud Interconnect and Cross-Cloud Interconnect, providing connectivity from on-premises or other cloud environments to Google Cloud. Available later this year.

• Unprecedented cluster scale: Build massive AI services with networking support for up to 30,000 GPUs per cluster in a non-blocking configuration, available in preview now.

• Zero-Trust RDMA security: Help secure your high-performance GPU and TPU traffic with our RDMA firewall, featuring dynamic enforcement policies for zero-trust networking. Available later this year.

• Accelerated GPU-to-GPU communication: Unleash up to 3.2Tbps of non-blocking GPU-to-GPU bandwidth with our high-throughput, low-latency RDMA networking, now generally available.

“Google Cloud plays a key role in our AI infrastructure, by supporting us to deliver high performance and secure AI experiences for our users at scale, while optimizing utilization of our resources.” – Xu Ning, Dir of Engineering, AI Platform, Snap, Inc.

The increasing complexity of AI inference, particularly when enterprises deploy multiple task-optimized models, presents significant networking challenges. The growing demand for AI capacity strains network infrastructure because efficiently routing data to GPU or TPU resources that are often distributed across regions requires high bandwidth and low latency. Furthermore, the introduction of gen AI applications and agents expands the attack surface, creating vulnerabilities for sensitive data leaks during inference, necessitating robust AI safety and security measures. To address these challenges, we are introducing GKE Inference Gateway, now in preview, which offers:

• Differentiated performance for gen AI applications without exorbitant serving costs. New capabilities in GKE Inference Gateway reduce serving costs by up to 30%, tail latency by up to 60% and increase throughput by up to 40% compared to other managed and open-source Kubernetes offerings based on internal benchmarks. Features in GKE Inference Gateway include intelligent load balancing based on model server metrics from Google Jetstream, NVIDIA, and vLLM, dynamic request routing, and efficient, dynamic LoRA fine-tuned models. 

• AI security and safety with powerful new integrations. Now, you can leverage the GKE Inference Gateway and Cloud Load Balancing alongside Model Armor, NVIDIA NeMo Guardrails, and Palo Alto Networks AI Runtime Security. This combined approach uses Service Extensions for comprehensive protection for your AI models, simplifying governance for platform engineering and security teams.

• Google Cloud Load Balancing optimizations for LLM inference, let you leverage NVIDIA GPU capacity across multiple cloud providers or on-prem infrastructure.

“Enterprises across industries are seeking full-stack, integrated infrastructure to deploy agentic AI securely and cost-effectively. By integrating NVIDIA inference software for real-time observability and NeMo Guardrails for robust security enforcement with GKE Inference Gateway, NVIDIA and Google Cloud are delivering advanced capabilities that enhance performance and reliability in AI deployments.” – Kari Briski, vice president of generative AI software for Enterprise, NVIDIA

Programmable global front end for web, media, and AI 

The Cross-Cloud Network Global Front End solution accelerates and secures the most demanding web, media, and now gen AI applications, regardless of where your backends are hosted, without having to expose your infrastructure to the internet. Today, we’re introducing new innovations for modern and gen AI apps:

• Edge programmability with Service Extensions: Unlock the power of the edge with open programmability through Service Extensions plugins, powered by WebAssembly (Wasm). Automate, extend, and customize your applications with over 60 plugin examples in Rust, C++, and Go. Cloud Load Balancing support is now generally available, and Cloud CDN support will follow later this year.

• Accelerated web performance: Deliver static and dynamic content at global scale with improved performance with Cloud CDN’s fast cache invalidation, and boost application performance for resumed connections with TLS 1.3 0-RTT. Both capabilities are now available in preview.

• End-to-end mTLS security: Strengthen your security posture with end-to-end mTLS, protecting your data from client to backend infrastructure via Cloud Load Balancing. Client-to-frontend mTLS was launched last year, and mTLS to backends is now in preview.

“Service Extensions plugins give us the ability to customize our web services by easily running custom code directly in the request/response path. Having an edge programmability solution based on open standards like WebAssembly with a large set of out-of-the-box examples allows our developers to quickly address custom requirements from the business.” – Justin Reid, Principal Engineer, Shopify 

Service-centric networking simplifies development

Whether you’re building cutting-edge gen AI applications or modernizing an existing system, service-centric architectures are essential for rapid iteration. As a pioneer of service-centric architectures, we are on a journey to help NetOps, DevOps, SecOps, and developer teams simplify service deployment and management. By abstracting away the complexities of the underlying networking and security layers, we empower developers to quickly deploy, update, and secure services across multiple applications. Today, we’re unveiling new innovations in automation, security, and scale with enhanced service-centric networking:

• Streamlined service discovery and management. App Hub integration simplifies producer-consumer interactions by automating service discovery and cataloging. Service health (coming later this year) enables resilient global services with network-driven cross-regional failover.

• Simplified multi-network, multi-service, multi-compute deployment. Later in 2025, you’ll be able to use Private Service Connect to publish multiple services within a single GKE cluster, making them natively accessible from non-peered GKE clusters, Cloud Run, or Service Mesh. 

“Our collaboration with Google has enabled us to streamline service discovery and helped to empower our developers to iterate faster and more efficiently.” – Jonathan Perry, Partner, Engineering, Goldman Sachs 

Protect modern and gen AI apps from evolving attacks

We’re witnessing a surge in sophisticated attacks: terabit-scale DDoS, DNS tunneling for data exfiltration, and the growing prevalence of AI-driven threats that evade traditional defenses. These cyber risks demand a fundamental shift in your approach to network security and underscore the need for advanced network security capabilities that extend beyond traditional perimeter defenses. Today, we’re announcing powerful network security enhancements that provide comprehensive protection for your distributed multi-cloud applications and internet-facing services.

Our strategy has three core pillars:

Secure the workload: Planet-scale DDoS protection, with up to 24x better threat efficacy

Protecting your distributed applications and internet-facing services against critical network attack vectors is paramount. Today, we’re introducing several key enhancements:

• DNS Armor: DNS traffic often lacks adequate monitoring, making it a prime target for data exfiltration. Attackers exploit this blind spot, using DNS tunneling, domain generation algorithms (DGA) and other sophisticated techniques to bypass traditional security controls. Powered by Infoblox Threat Defense with visibility into 70 billion DNS events daily, DNS Armor detects these DNS-based data exfiltration attacks. Available in preview later this year.

• Enhanced security posture enforcement: Strengthen your security posture with consistent org-wide protection using new hierarchical policies for Cloud Armor. Enforce granular protection independent of your network architecture with new network types and new firewall tags for Cloud NGFW Hierarchical firewall policies, coming this quarter in preview.

• In 2024, we launched Cloud NGFW Enterprise with up to 24x higher efficacy than other major public clouds. We are continuing to improve Cloud NGFW with new layer 7 domain filtering, which will allow firewall administrators to monitor and control outbound web traffic to only allowed destinations, coming later in 2025.

“We use Cloud NGFW and Cloud Armor to protect our critical applications and websites in Google Cloud. The new Network Security innovations announced at Next will help us improve protection for our users and simplify how we manage network security.” – Jason Jones, Sr. Director, Security Engineering, UKG

Secure the data: Introducing inline network DLP

In today’s data-driven world, your enterprise’s intellectual property is its most valuable asset. But ensuring its security and compliance can be complex. We understand the need for robust, yet streamlined, data loss prevention (DLP) across both data at-rest and data in-transit. Our upcoming inline network DLP for Secure Web Proxy and Application Load Balancer provides real-time protection for sensitive data-in-transit via integration with third-party (Symantec DLP) solutions using Service Extensions. In preview this quarter, inline network DLP helps you safeguard critical data and maintain compliance, without sacrificing performance or agility.

Open security ecosystem: Third-party security insertion

We give you the flexibility to choose your preferred security solutions, tailoring protection to your specific needs. We are excited to expand our security partner ecosystem with deeper integrations. Recently, we announced that you can insert partner network services or virtual appliances with Google Cloud workloads via Network Security Integration. Now generally available, this helps you maintain consistent policies across hybrid and multi-cloud environments without changing your routing policies or network architecture. 

Plus, to broaden our web and API protection ecosystem, we’ve worked with Imperva to integrate Imperva Application Security with Cloud Load Balancing, also via Service Extensions, and it’s now available in the Google Cloud Marketplace.

Cloud WAN: the enterprise backbone for the AI era 

Connecting a modern business is incredibly complex. Customers have to deal with many different networks and security architectures, and must make difficult choices to balance reliability, application speeds, and cost. This can lead to complex, custom solutions that are hard to manage, weaken security postures, and often don’t deliver the best results. Cloud WAN, our newest Cross-Cloud Network solution, is a fully managed, reliable and secure enterprise backbone to transform enterprise WAN architectures and address these challenges. 

Cloud WAN delivers significant advantages : 

• Cloud WAN provides up to a 40% savings in total cost of ownership (TCO) over a customer-managed WAN solution leveraging colocation facilities¹

• Global reach and performance via Google’s expansive backbone network with 99.99% reliability

• Cross-Cloud Network up to 40% improved performance compared to the public internet²

• An open, flexible and tightly integrated ecosystem with major SD-WAN and security vendors 

For more details please read the full announcement here.

A network to deliver the AI era

Our cloud networking products and solutions give you the ability to connect, simplify, modernize and secure your organizations across the planet. With these new innovations — plus the new Cloud WAN — we continue to provide you with the flexibility to adapt to new technologies, services, applications and locations, all with the agility required for the AI era.

For more on our Google Cloud Next 2025 announcements, you can watch our Cross-Cloud Network innovations session, and check out the many great networking breakout sessions.

<sup>1. Architecture includes SD-WAN and 3rd party firewalls, and compares a customer-managed WAN using multi-site colocation facilities to a WAN managed and hosted by Google Cloud.
2. During testing, network latency was more than 40% lower when traffic to a target traveled over the Cross-Cloud Network compared to when traffic to the same target traveled across the public internet.</sup>

Generative AI continues to capture our imagination and promises to transform every industry. Its transformative potential hinges on integrating powerful models like Gemini, with the most contextually-relevant enterprise data.

Google Cloud is leading this transformation, not only by delivering the gen AI technologies themselves, but also by embedding those technologies directly into operational databases. For example, since introducing AlloyDB AI with vector search, we’ve expanded the capability across all our other database services — Bigtable, Cloud SQL, Firestore, Memorystore, and Spanner.

Today, operational databases are powering new AI agents and multimodal applications, and are enhancing existing applications with AI capabilities. That’s why we’re committed to delivering a truly innovative and unified data and AI foundation, helping customers redefine their industries, tackle massive data challenges, and drive groundbreaking innovation.

Today, at Google Cloud Next, we’re unveiling new database capabilities, including:

1. Generative AI capabilities in AlloyDB and agentic programming with MCP

2. MongoDB compatibility in Firestore

3. Expanded Oracle services, and a SQL Server modernization solution

New generative AI capabilities in AlloyDB and agentic programming with MCP

Agentic workflows are emerging as a primary architectural pattern for AI-based applications, and databases will be a primary component in this. The best agents will be the ones that can use real-time data to enable high quality, agentic decisions and actions — bolstering the quality of reasoning. To support this, we’re investing in AlloyDB AI capabilities to make it easier for developers to build intelligent agents and applications. 

First, we’re enabling Google Agentspace to search structured data in AlloyDB. Agentspace brings together Gemini’s advanced reasoning with Google-quality search and enterprise data. Now you can activate all your data stored in AlloyDB, enabling you to combine real-time, structured, and unstructured data in creative ways.

Second, we know that developers are looking to enable more flexible interfaces to their databases, while staying secure and accurate. That’s why last year we announced natural language support in AlloyDB to help developers build applications which accurately query data with natural language, just like they do with SQL.

We’ve been iterating on this, and today, we’re launching the next-generation of AlloyDB natural language. This technology lets you query structured data in AlloyDB securely and accurately, enabling natural language text modality in apps. 

AlloyDB AI natural language goes beyond interpreting database metadata, using supplied context and interactive intent clarification when querying the database. You can define which data can be accessed by natural language queries by using AlloyDB’s parameterized secure views, which provide an extra layer of security for agents and gen AI apps.

Third, building intelligent apps requires powerful vector search. We’ve seen AlloyDB’s vector search adoption increase nearly seven times since the launch of the state-of-the-art Scalable Nearest Neighbor (ScaNN) for AlloyDB index in 2024. Along with enhancements to ScaNN indexing in AlloyDB, we now offer optimized SQL functionality spanning vector search and structured filters and joins. 

With these innovations, AlloyDB’s ScaNN index offers up to 10 times faster filtered vector search queries compared to the hierarchical navigable small world (HNSW) index in standard PostgreSQL.

Fourth, in partnership with Vertex AI and Google DeepMind, we are introducing three new models in AlloyDB AI — one that improves the relevance of vector search results using cross attention reranking; a multimodal embeddings model that supports text, images, and videos’ and a new state-of-the-art Gemini Embedding text model. As a result you can easily add intelligence into your apps across multiple modalities such as images and text, with high efficiency and accuracy.

Fifth, to further infuse AI experiences into apps, we’re introducing AlloyDB AI query engine. AI query engine enables developers to freely use natural language expressions and constructs right within their SQL queries in a natural way. Developers can now use free text questions like “find family-friendly hotels in Orlando,” that require real-world data, including images and descriptions, and directly embed them in their SQL queries. 

At the heart of this are industry-leading foundation model-powered semantic operators that run side-by-side with traditional relational operators in the AlloyDB AI query engine.

Many of these AlloyDB capabilities are available in preview, and you can sign up to use them here today.

“At Target, we used AlloyDB to improve our online search experience. We used the ability to combine our structured and unstructured data to enhance the accuracy of natural language search queries by 20%!” Visagan Subburayalu, VP of Infrastructure & Cybersecurity, Target

Finally, we are announcing that MCP Toolbox for Databases (formerly Gen AI Toolbox for Databases) now supports Model Context Protocol (MCP). MCP enables seamless connections between AI agents and enterprise databases, replacing custom code that would otherwise be required. MCP Toolbox for Databases is an open source (Apache 2.0) server that supports multiple databases including PostgreSQL, MySQL, AlloyDB, Spanner, Cloud SQL (for PostgreSQL, MySQL, and SQL Server), Neo4j and Dgraph. It offers simplified development with reduced boilerplate code, enhanced security through OAuth2 and OIDC, and end-to-end observability with OpenTelemetry integration. We have open-sourced the code, so you are free to add whatever databases you choose.

Big step forward for document databases with MongoDB compatibility in Firestore

Developers love the agility of the popular MongoDB API and query language to store and query semi-structured JSON data, including the ability to use the open-source ecosystem of MongoDB integrations. Consequently, customers are looking for increased choice in how to build and deploy these workloads.

Today, we’re announcing the preview of Firestore with MongoDB compatibility. Built from the ground up by Google Cloud, it provides developers with additional choice for their demanding document database workloads. MongoDB API compatibility has been a highly-requested capability from Firestore’s existing community of over 600,000 monthly active developers. 

With this launch, Firestore developers can now take advantage of MongoDB’s API portability along with some of the Firestore capabilities they have come to depend on. These capabilities include multi-region replication with strong consistency, virtually unlimited scalability, industry-leading availability of up to 99.999% SLA, and single-digit milliseconds read latency performance, without having to worry about managing the underlying database infrastructure. 

In addition, developers can use their existing MongoDB application code, drivers, and integrations with the Firestore service. Firestore with MongoDB compatibility offers a customer-friendly serverless pricing model, with no up-front commitments required, and customers only pay for what they use. 

We’re committed to our partner ecosystem and will continue to support MongoDB Atlas in our Google Cloud Marketplace. This launch simply offers developers an additional choice for building their applications. You can get started here today.

“After migrating to Firestore, we improved developer productivity by 55%, observed better service reliability, and have been able to seamlessly scale to over 250,000 requests per second and 30 billion documents. Because Firestore is completely serverless and provides virtually unlimited scalability, we no longer have to worry about managing our underlying database infrastructure – liberating us from database DevOps. This has enabled us to focus on product innovations that matter to our customers,” said Karan Agarwal, director of engineering, HighLevel.

Expanded range of Oracle services, and a SQL Server modernization solution

Last year we announced Oracle Database@Google Cloud with availability in four global regions, enabling customers to migrate Oracle workloads, and modernize them with Google’s industry-leading data and AI capabilities such as BigQuery, Vertex AI, and Gemini foundation models. 

Today we’re announcing support for Oracle Base Database Service, offering a flexible and controllable way to run Oracle Databases in the cloud. We’re also announcing general availability of Oracle Exadata X11M, bringing the latest generation of the Oracle Exadata platform to Google Cloud and offering additional enterprise-ready capabilities, including customer managed encryption keys (CMEK). 

We’re continuing to invest in global infrastructure for Oracle, and these services are being deployed natively in 20 Google Cloud locations.

Building on this foundation, customers can now accelerate the development of cutting-edge agentic applications by integrating the power of the Google ecosystem with their business data residing in Oracle Database@Google Cloud. Learn more in this partner blog.

“Banco Actinver is committed to providing innovative financial solutions to our clients. By combining the security and performance of Oracle Database with Google Cloud’s data analytics and AI tools, we’re gaining deeper insights into market trends, enhancing our services, and delivering personalized experiences to our customers.” Jorge Fernandez, CIO, Banco Actinver

We know that many organizations are looking to reduce their dependence on SQL Server in order to eliminate its high costs and unfriendly licensing. Customers want automated ways to move their workloads to modern, AI-enabled databases.

Today, we’re announcing that Database Migration Service (DMS) now supports SQL Server to PostgreSQL migrations for Cloud SQL and AlloyDB. 

These new capabilities support the migration of both self-managed and cloud-managed SQL Server offerings, as well as a wide range of SQL Server editions and versions to allow you to fully execute on your database modernization strategy.

DMS offers online data migration and features a schema and code conversion engine that utilizes a combination of algorithmic and a specially-trained Gemini model, all designed to automate the most difficult migration steps, like converting Transact-SQL code and SQL Server-specific data types like DATETIME to their PostgreSQL equivalents.

But that’s not all, additional databases announcements at Next include:

1. Cloud SQL and AlloyDB are available on C4A instances. C4A instances are based on Google Axion Processors, our first Arm-based CPUs, custom-built for the cloud. These instances deliver nearly 50% better price-performance compared to N series machines, and up to two times better throughput than Amazon’s equivalent Graviton4-based offerings. Learn more here. 

2. Database Center, our AI-powered unified fleet management solution, is generally available and supports every database in our portfolio. This release delivers richer metrics and actionable recommendations, empowering users to optimize performance and reliability for their database fleet.

3. Spanner vector search is now generally available, designed to work with our SQL, Graph, Key-Value, and Full-Text Search modalities to address the most demanding AI workloads at virtually unlimited scale. 

4. Graph Visualization for Spanner, is now generally available. It allows users to visually explore valuable information from graph data. You can drill into specific nodes and relationships, filtering and highlighting relevant data subsets, and exploring the graph through intuitive navigation.

5. Aiven for AlloyDB Omni, is now generally available. This is a fully-managed AlloyDB Omni service, from our partner Aiven, running on AWS, Azure, and Google Cloud. 

6. Bigtable continuous materialized views, in preview, delivers an enhanced developer experience that simplifies real-time updates for modern applications that rely on immediate reporting and insights.

7. Memorystore for Valkey is now generally available, it supports 7.2 and 8.0 engine versions. Memorystore for Valkey 8.0 achieves up to 2x Queries Per Second (QPS) at microseconds latency when compared to Memorystore for Redis Cluster and provides optimized memory efficiency and better reliability for customers.

The future of data with Google Cloud

We’re excited to keep innovating on your behalf. At Google Cloud, we offer an intelligent, unified, and open data platform to support your generative AI journey every step of the way.

Learn more about Google Cloud databases and start a free trial for Cloud SQL, AlloyDB, and Spanner.

For over 25 years, Google has pushed the boundaries of the network, with innovations that connect billions of users around the world to essential services like Gmail, YouTube, and Search. At the foundation lies Google’s vast backbone network. With 202 points of presence (PoPs), powered by over 2 million miles of fiber, 33 subsea cables, and backed by a 99.99% reliability SLA, Google’s network provides a robust and resilient global platform. 

That same planet-scale network infrastructure powers Google Cloud and our Cross-Cloud Network solutions. And today we are announcing that Google’s global network is available for all businesses and governments to use with our new Cloud WAN solution.

Cloud WAN is a fully managed, reliable, and secure enterprise backbone to transform enterprise wide area network (WAN) architectures. It leverages Google’s planet-scale network, which is optimized for application performance. Cloud WAN provides up to 40% faster performance¹ compared to the public internet, and up to a 40% savings² in total cost of ownership (TCO) over a customer-managed WAN solution.

The evolution of the enterprise WAN

Traditionally, enterprises relied heavily on multiprotocol label switching (MPLS) networks for secure and reliable site-to-site connectivity — a high-cost approach. However, the rapid adoption of SaaS and cloud applications necessitated a transformation, leading to the emergence of SD-WAN with direct internet access (DIA), a lower-cost alternative that leveraged the internet. Then, to enhance application performance, enterprises built colocation-based cloud on-ramps, which, while improving latency, introduced complexity and costs. This evolution led to a proliferation of security stacks, with a mix of security service edge (SSE) and self-managed appliances, creating an inconsistent security posture. 

Consequently, enterprise connectivity has become inherently complex and difficult to manage, marked by diverse networks with fragmented security, and requiring a constant balancing of reliability, speed, and cost. 

The rapid emergence of AI has placed additional demands on enterprise networks. AI-based applications require a highly distributed infrastructure that is often spread across different clouds and on-premises data centers, and need a global network that can scale massively, provide robust security and privacy, and make efficient use of resources, all while remaining cost-effective.

To address these needs, Cloud WAN offers a unified enterprise network solution for securely and reliably connecting across any enterprise site, application and user, while helping to ensure optimal performance and reducing costs.

Cloud WAN enables two key use cases: providing high-performance connectivity between geographically dispersed data centers, and connecting branch and campus environments over our Premium Tier network.

Use case 1: High performance, cross-region connectivity 

Large, global customers with extensive data center networks need to be able to move large amounts of data reliably. Cloud WAN offers flexible connectivity options for linking geographically dispersed data centers, providing a modern alternative to traditional solutions that have limited capacity, high cost of operations, and lower reliability.

This use case is enabled by these core capabilities:

Cloud Interconnect provides private and dedicated low-latency connections between Google Cloud regions and on-prem data centers, allowing you to connect your data center to Google Cloud from over 159 locations around the globe.

Cross-Cloud Interconnect enables multicloud connectivity between Google Cloud and other public cloud environments, including AWS, Azure and OCI, and is available in 21 locations. 

Now, with Cross-Site Interconnect, we’re enabling dedicated point-to-point layer 2 private 10/100G connectivity optimized to connect applications between data centers, and that is designed for enterprises, government agencies, and telecom operators. With Cross-Site Interconnect, Google Cloud becomes the first major cloud provider to offer transparent layer 2 connectivity over its network. Currently in preview in select countries, we’ll expand it to additional edge locations in the coming year.

Key benefits of Cross-Site Interconnect include: 

• Enterprise-grade performance: With an SLA and built-in redundancy, and Google’s high performance global network, Cross-Site Interconnect is a high-bandwidth (10/100G) data center interconnect solution.

• Global reach and capacity: Cross-Site Interconnect runs over Google Cloud’s global backbone, a network of over 2 million miles of fiber, with international capacity for trans-Atlantic and trans-Pacific networks.

• Optimized for risk and cost: Cross-Site Interconnect leverages redundant fiber infrastructure to dynamically reroute network traffic around physical network and optical-fiber failures, helping to maintain consistent service availability. 

Note: Google Cloud additionally supports scalable layer 3 routing solutions with Network Connectivity Center (NCC). Please see the next section for details. 

“Having a scalable and reliable global network is critical as we continue to expand our research efforts and host performance sensitive workloads on Google Cloud. We look forward to leveraging the stability and reliability we have seen with Google’s Cloud WAN to meet the growing needs in our hybrid networks.” – Chris Dee, Head of Cloud Platform Engineering, Citadel Securities

Use case 2: Migrate branch and campus networks

Google’s Premium Tier network serves as a powerful cloud on-ramp, helping to securely connect branch offices and campuses to public cloud resources, SaaS applications, and the internet. Cloud WAN is a unified, fully managed solution that extends the performance and reliability of Google’s trusted network directly to enterprises, enabling on-demand, any-to-any connectivity with inherent security and a lower TCO. Furthermore, our open yet tightly integrated ecosystem empowers enterprises with the choice of best-in-class security and network services, fostering flexibility and tailored solutions.

Premium Tier network delivers up to 40% improved performance

The foundation of this performance is Google’s unparalleled backbone network which provides a robust and resilient platform for your applications.

• Google’s Premium Tier network service is designed to deliver optimal application performance. It ensures that internet-based traffic targeting Cloud WAN enters and exits Google’s high-performance network at the geographically closest PoP. This minimizes network hops, driving lower latency and a more consistent user experience.

• With over 5700 direct peering connections and reachability to 60K+ autonomous system numbers (ASNs), Google ranks first among cloud providers and 6th in global peering. This extensive peering network helps ensure efficient traffic exchange with other networks, further enhancing application performance.

• Our Verified Peering Provider program allows customers to select ISPs for maximum availability. Now open to all ISPs globally, the program provides redundant, diverse connectivity. 

• Dedicated fiber access in partnership with Lumen Technologies (available in select locations in 2025), enables last-mile handoff to Cloud WAN from customer-operated locations including data centers, branches, warehouses, and airports. 

• BT is enabling their Global Fabric network-as-a-service (NaaS) offering to provide direct connectivity for BT customers to Google Cloud using Cloud WAN

Cloud WAN’s architecture offers flexible options to address modern enterprises’ diverse connectivity requirements. Network Connectivity Center acts as a centralized hub for enterprises to connect their branches using Cloud VPN or their preferred SD-WAN solution, while campuses and data centers can use Cloud Interconnect for high-bandwidth (10G, 100G) links. To further extend the reach and reliability of enterprise networks, site-to-site data-transfer with Network Connectivity Center is now available in 20 countries across the globe.

“By bridging the edge-cloud gap, Cloud WAN is enhancing our application performance by up to 40%, reducing costs, and providing resilient, scalable support and foundation for our end-to-end digitalization.” – Ralf Huebenthal, Global Head of IT Platforms, Nestle

By combining these powerful network capabilities, Cloud WAN delivers a significant improvement in application experience, empowering businesses to provide their users with the performance they expect.

Cloud WAN lowers TCO by up to 40%
Compared with a customer-managed WAN solution, Cloud WAN provides up to a 40% savings in TCO², offering the flexibility of both usage-based and fixed-price options to suit various needs, and significantly driving down fixed, upfront costs. For organizations that perform high-bandwidth internet data transfers, we are extending a fixed-price model for Cloud Interconnect that will be generally available this quarter. Cloud WAN allows enterprises to consolidate their carrier-neutral facility deployments to fewer cloud regions, reducing costs. 

A tightly integrated, open ecosystem for flexibility and choice
Cloud WAN has an open, flexible, and tightly integrated ecosystem, providing enterprises with the tools needed to build robust and adaptable networks. Cloud WAN offers easy integration of network and security services from a broad ISV ecosystem:

• Strong SD-WAN partner ecosystem, featuring leading vendors including Cisco, Fortinet, Juniper Networks, Netskope, Velocloud, and others. Enterprises can choose to self-manage or consume managed overlay connectivity with our services partners.

• Cloud WAN using NCC Gateway (preview in Q2), the first major cloud solution to offer managed integration of security service edge (SSE) for users accessing private and public applications. SSE offerings are available from Broadcom, Menlo Security, and Palo Alto Networks.

• Flexible security options letting you use either Cloud Next Generation Firewall (NGFW), or that integrate NGFW virtual appliances from Check Point, Fortinet, and Palo Alto Networks via Network Security Integration.

You can also extend the benefits of Cloud WAN with partner solutions to build agile, secure, and automated branch networks, facilitating rapid deployment of new branches and services enabled with: 

• Infoblox Universal DDI modernizes and simplifies branch networks by delivering infrastructure-free, cloud-based Domain Name Service (DNS), Dynamic Host Configuration Protocol (DHCP), and IP Address Management (IPAM). An optional DNS security component protects against phishing and malware, ensuring your users connect safely to critical applications.

• Juniper Networks Mist provides AI-driven campus and branch transformation, including comprehensive insight and automation to optimize  user experiences. This integration includes cloud-delivered critical network services like Wi-Fi, wired access, SD-WAN, network access control (NAC), indoor location services, and IoT analytics.

• Integration with Google Distributed Cloud (GDC) in a connected configuration, extending Google Cloud’s computing benefits to the edge for latency-sensitive applications. GDC is a fully managed platform supporting containerized and VM-based applications, with automated connectivity back to Google Cloud regions using Cloud WAN. 

Partner-managed services 

Enterprises who already rely on managed services providers for their enterprise backbone needs can continue to work with their preferred partners when migrating to Cloud WAN. And for enterprises that want help migrating, deploying and operating Cloud WAN, they can engage with one of our global system integrator (GSI) partners. At launch, Cloud WAN partners include Accenture, HCLTech, and Wipro, who now offer services architecture, design, migration and ongoing management services.

Learn more

Cloud WAN stands to revolutionize how enterprises connect and secure their global infrastructure. Offering simplicity, high performance, a wealth of connectivity and security service choices, not to mention significant cost savings, Cloud WAN lets you focus on innovation and growth in the cloud and beyond. 

You can learn more about Cloud WAN at Next ‘25 and on the Cross-Cloud Network solution page. Plus check out the many other great networking breakout sessions at Google Cloud Next 2025.

<sup>1. During testing, network latency was more than 40% lower when traffic to a target traveled over the Cross-Cloud Network compared to when traffic to the same target traveled across the public internet.
2. Architecture includes SD-WAN and 3rd party firewalls, and compares a customer-managed WAN using multi-site colocation facilities to a WAN managed and hosted by Google Cloud.</sup>

The age of AI is now. In fact, the global AI infrastructure market is on track to increase to more than $200 billion by 2028. 

However, working with massive data, intricate models, and relentless iterations isn’t easy, and adapting to this new era can be daunting. Platform engineering and infrastructure teams that have invested in Kubernetes may wonder: “After years of building expertise in container orchestration to operate production workloads at scale, how do I enable this next generation of AI workloads?”

The good news? You don’t need to start from scratch. You’re well on your way — your Kubernetes skills and investments aren’t just relevant, they’re your AI superpower.

Today, at Google Cloud Next, we’re announcing significant improvements to Google Kubernetes Engine (GKE) to help platform teams succeed with AI:

• Cluster Director for GKE, now generally available, lets you deploy and manage large clusters of accelerated VMs with compute, storage, and networking — all operating as a single unit.
• GKE Inference Quickstart, now in public preview, which simplifies the selection of infrastructure and deployment of AI models, while delivering benchmarked performance characteristics.
• GKE Inference Gateway, now in public preview, provides intelligent routing and load balancing for AI inference on GKE.
• A new container-optimized compute platform is rolling out on GKE Autopilot today, and in Q3, Autopilot’s compute platform will be made available to standard GKE clusters.
• Gemini Cloud Assist Investigations, now in private preview, helps with GKE troubleshooting, decreasing the time it takes to understand the root cause and resolve issues.
• With a built-in partnership with Anyscale, RayTurbo on GKE will launch later this year to deliver superior GPU/TPU performance, rapid cluster startup, and robust autoscaling.

Read on for more details about these announcements.

Scale your AI workloads with Cluster Director for GKE

As AI models grow in size and demand more machines for compute processing, platform teams need to deliver new architectures to deploy models across multiple hosts and operate massive clusters of GPUs and TPUs as a single unit. Without these capabilities, customers often struggle to complete large training jobs and to deliver the inter-machine performance they need for AI.

To handle these scaling challenges, our supercomputing services, Cluster Director for GKE (formerly Hypercompute Cluster), is now generally available. With Cluster Director for GKE, you can deploy and manage large clusters of accelerated VMs with compute, storage, and networking — all operating as a single unit. It delivers exceptionally high performance and resilience for large distributed workloads by automatically repairing faulty clusters based on their bill of health.

One of the best things about Cluster Director for GKE is that you can orchestrate all of this through standard Kubernetes APIs, and ecosystem tooling. There are no new platforms — just new capabilities on the platform you already know and love. You can use GKE node labels to:

• Schedule pods based on network topology to maximize efficiency and minimize network hops.

• Report and replace faulty nodes by gracefully evicting workloads from the node and automatically replacing them with spare capacity within your co-located zone.

• Manage host maintenance so you can manually start host maintenance from GKE or use maintenance information while scheduling your workloads.

To get started with Cluster Director for GKE, use configurable blueprints in Cluster Toolkit or Accelerated Processing Kit (XPK), a command-line tool that requires no prior Kubernetes knowledge.

Make your apps smarter with inferencing on GKE

We are seeing a clear trend in the age of AI: amazing innovation is happening where traditional compute interacts with neural networks — otherwise known as “inference.” Companies operating at the cutting edge of Kubernetes and AI, like LiveX and Moloco, run AI inference on GKE.

Customers and platform teams deploying AI inference on Kubernetes tell us they face two key challenges:

1. Balancing performance and cost: Tuning accelerators to meet the right performance targets without overprovisioning requires extensive knowledge of Kubernetes, AI models, GPU/TPU accelerators, and specific inferencing metrics like Time To First Token (TTFT).

2. Model-aware load balancing: With AI models, response length is often highly variable from one request to another, so response latency varies widely. This means traditional load balancing techniques like round-robin can break down, exacerbating latency and underutilizing accelerator resources.

To address these challenges, we’re introducing new AI inference capabilities in GKE:

1. A new GKE Inference Quickstart, now in public preview, lets you pick an AI model and then provides a set of benchmarked profiles to choose from. The profiles include infrastructure configuration, GPU/TPU accelerator configuration, and Kubernetes resources needed to match a set of AI performance characteristics like TTFT.

2. GKE Inference Gateway, now in public preview, reduces serving costs up to 30%, tail latency by up to 60%, and increases throughput by up to 40%. Customers get a model-aware gateway optimized for intelligent routing and load balancing, including advanced features for routing to different model versions.

The best solutions to complex problems meet you where you are and point you to what’s next. The combination of the GKE Inference Quickstart and GKE Inference Gateway do just that.

Optimize workloads with GKE Autopilot

Optimizing cloud usage and cost savings continues to be a top priority for both Google Cloud and for cloud users, with 71% naming it as their top initiative for the year. If you’re running web and API servers, queue processors, CI/CD agents, or other common workloads, you’re likely over-provisioning some resources to make your apps more responsive. GKE customers often request more compute resources than they use, leading to underutilization and unnecessary costs. 

In 2021, we launched GKE Autopilot to combat overprovisioning. Autopilot dramatically simplifies Kubernetes cluster operations and enhances resource efficiency. More and more customers, including Toyota and Contextual AI, are turning to Autopilot for critical workloads. In fact, 30% of active GKE clusters created in 2024 were created in Autopilot mode.

Today, we are announcing new performance improvements to GKE Autopilot, including faster pod scheduling, scaling reaction time, and capacity right-sizing — all made possible by unique hardware capabilities only available on Google Cloud. With Autopilot, your cluster capacity is always right-sized, allowing you to serve more traffic with the same resources or existing traffic with fewer resources.

Currently, Autopilot consists of a best-practice cluster configuration and a container-optimized compute platform that automatically right-sizes capacity to match your workloads. Many customers have told us that they want to right-size capacity on their existing clusters without having to use a specific cluster configuration. To help, starting in Q3, Autopilot’s container-optimized compute platform will also be available to standard GKE clusters, without requiring a specific cluster configuration.

Save time with Gemini Cloud Assist

Nothing slows down the pace of innovation like having to diagnose and debug a problem in your application. Gemini Cloud Assist provides AI-powered assistance across the application lifecycle, and we’re unveiling the private preview of Gemini Cloud Assist Investigations, which helps you understand root cause and resolve issues faster. 

The best part? It’s all available right from the GKE console, so you can spend less time troubleshooting and more time innovating. Sign up for the private preview to be able to: 

• Diagnose pod and cluster issues from the GKE console — even across other Google Cloud services such as nodes, IAM, or load balancers.

• See observations from logs and errors across multiple GKE services, controllers, pods, and underlying nodes.

Kubernetes is the open infrastructure platform for AI

For organizations looking for a comprehensive machine learning platform, we recommend Vertex AI, a unified AI development platform for building and using generative AI on Google Cloud. With access to Vertex AI Studio, Agent Builder, and over 200 models in Model Garden — plus the ability to call it from GKE — it’s a great option if you’re looking for an easy-to-use solution.

Over the last decade, Kubernetes has earned its spot as the de-facto standard for hosting cloud-native applications and microservices. Today, organizations that need deep control over their infrastructure are once again turning to Kubernetes to build their AI training and inference platforms. In fact, IBM, Meta, NVIDIA, and Spotify all use Kubernetes for their AI/ML workloads.

To make Kubernetes an even better platform for AI, we’re proud to lead and contribute alongside these companies (and more with the Cloud Native Computing Foundation) to create exciting open source innovations:

• Built with Intel, NVIDIA, and more, Dynamic Resource Allocation simplifies and automates hardware allocation and scheduling to pods and workloads.

• Developed in conjunction with Apple, Red Hat, and others, Kueue and JobSet provide powerful AI training orchestration, streamline job management, and optimize accelerator utilization.

• Partnering with DaoCloud, we built LeaderWorkerSet, which enables deployment and management of large, multi-host AI inference models via a Kubernetes-native API.

Empower data scientists and AI/ML engineers with Ray on GKE

Platform teams have historically relied on Kubernetes and GKE to serve the needs of software engineers building microservices and related cloud-native applications. With increased AI usage, these same platform teams now need to serve a new user base: data scientists and AI/ML engineers. However, most data scientists and AI/ML engineers aren’t familiar with Kubernetes and need a simpler, more approachable way of interacting with distributed infrastructure. 

To curb the steep learning curve, many organizations turn to Ray, an open-source framework that provides an easy way for AI/ML engineers to develop Python code on their laptops and then scale that same code elastically across a Kubernetes cluster. 

We’re committed to making Kubernetes the best platform for using Ray, and have been working closely with Anyscale, the creators of Ray, to optimize open-source Ray for Kubernetes. Today, in partnership with Anyscale, we’re announcing RayTurbo on GKE, an optimized version of open-source Ray. RayTurbo delivers 4.5x faster data processing and requires 50% fewer nodes for serving. On GKE, RayTurbo will take advantage of faster GKE startup times, high-performance storage for model weights, TPUs, and better resource efficiency with dynamic nodes and superior pod scalability. RayTurbo on GKE will launch later this year.

Better together: GKE and AI

AI introduces new challenges for platform teams, but we’re confident that the technology and products that you already use — Kubernetes and GKE — can help you tackle them. With the right foundation, platform teams can expand their scope to support data scientists and AI/ML engineers in addition to software engineers. 

This confidence is grounded in experience. At Google, we use GKE to power our leading AI services — including Vertex AI — at scale, relying on the same technologies and best practices that we’re sharing with you today. 

Ready to start using Kubernetes for the next generation of AI workloads? We look forward to watching you succeed on GKE.

From answering search queries, to streaming YouTube videos, to handling the most demanding cloud workloads, for over 25 years, we’ve been relentlessly pushing the boundaries of network technology, building a global infrastructure that powers Google and Google Cloud for billions of users and enterprise customers globally. We now stand at another pivotal moment, driven by the transformative power of AI, and our network is once again evolving to meet the challenges and opportunities of this new era. 

Here’s a look behind the scenes at the evolution of our global network, from enabling the early days of web search, to today, powering demanding AI workloads to bring AI’s benefits to everyone — people and businesses alike.

Our network’s evolution

There have been several fundamental inflection points for the Google network over the last 25 years, leading to three distinct networking eras:

Internet era: Our journey began in the internet era, when we primarily focused on offering our users across the globe a consistently high-quality experience in terms of reliability and latency — whether they were using Search, Maps, or Gmail. Key innovations included the B2 network; Bandwidth Enforcer (BwE); B4, our first fully software-defined backbone; our Orion software-defined network (SDN) controller; and our petabit-scale SDN data-center fabric, Jupiter. 

Streaming era: With the advent of YouTube and similar services, streaming video became a significant portion of global internet traffic — a trend that continues even today. We adapted our network to deliver low-jitter and high-quality video around the world through technologies such as Google Global Cache, Espresso, QUIC, and TCP BBR. 

Cloud era: The rise of cloud computing demanded greater resiliency, multi-tenancy, and security, which inspired innovations such as Andromeda, gRPC, PSP, and Swift.

Alongside technology innovations, our network footprint had to scale continuously to reach every Google user and customer with a consistent, high-quality experience. Today, this network spans over 2 million miles of lit fiber, including 33 subsea cable investments, with 202 network edge locations and more than 3,000 media content delivery network (CDN) locations across the globe. It connects 42 Google Cloud regions and 127 zones. We are also the most deeply peered cloud service provider network in the world.

AI is driving unprecedented network demands 

As Sundar noted in his Google I/O 2024 keynote, we’ve been AI-first in our approach for more than a decade, investing in and innovating at every layer of the stack. From research and products to infrastructure — our global network fuels these AI innovations and brings them to you wherever you are in the world. All 15 of our half-billion-user products — including seven with 2 billion users — are powered by our Gemini model, and all of them rely on the Google global network to bring us closer to our ultimate goal: making AI helpful for everyone. We take this responsibility very seriously. 

The AI era presents unique challenges that require a fundamental rethinking of our network architecture from four key perspectives:

• The wide area network (WAN) is the new local area network (LAN): In the AI era, we train the largest of our foundation models across multiple campuses and even multiple metros to pool together large numbers of TPUs. The need for scalability has never been this acute, both for Gemini and for our customers building foundation models on Google Cloud infrastructure. Moreover, these ML applications have unique traffic patterns, such as highly bursty elephant flows. Understanding and managing these flows is critical for efficient network performance. 

• AI demands zero impact from any outages: AI foundation model training, fine-tuning and inferencing are intensive processes that rely on valuable GPU/TPU resources, and a prolonged outage can be very disruptive to them. In other words, network disruptions are simply unacceptable — our customers expect always-on connected network capacity.  

• A heightened need for security and control: AI models and the data they are trained on must be protected to ensure their integrity. In addition, there are evolving compliance requirements for AI models from different regions and for data in transit.

• Operational excellence: From creating site reliability engineering (SRE) principles and leveraging AI/ML innovations in operations, to finding failure root causes using ML, we’re always exploring new ways to deliver excellence in our network operations. Simultaneously, the challenges of costs and complexity associated with linear scaling have pushed us to seek solutions that are efficient and sustainable for our customers.

New network design principles and innovations

To address these challenges, we’ve reimagined our next-generation network from the ground up, establishing four new design principles.

1. Exponential scalability: Our network needs the ability and agility to handle massive amounts of data and traffic, especially in key regions serving AI traffic. The need for scalability has never been greater. In the AI era, the WAN is the new LAN and the continent is the data center. 
2. Beyond-9s reliability: The industry has traditionally understood reliability in terms of “3 9s,” “4-9s” or “5-9s” of availability. Increasingly, that is simply not enough, as long-tail events, well within the x-9s specifications, matter as much as the average reliability of the network. Our users and customers expect deterministic performance, a limited impact radius for incidents, and proactive and ultra-fast mitigation. We are embarking on a journey to go “beyond 9s.” 
3. Intent-driven programmability: Billions of people use our network. They have unique requirements for security, compliance, resiliency, performance, and efficiency. To address all these requirements, we need a fully intent-driven, highly programmable network. 
4. Autonomous network: Automation and zero-touch have been buzzwords for the last decade. To support the next decade’s demands, we need autonomous networks that can run at scale 24×7 with minimal human intervention.

Guided by these four design principles, we have built our next-generation global network by making  foundational networking advancements.

Multi-shard network: We are moving beyond traditional vertical scaling limitations to elastic, horizontal scalability with our multi-shard network architecture. Each network shard is independent and enables horizontal scaling; not only can we scale the network within a shard, but we can scale the number of shards in the network. This allows for swift and substantial WAN bandwidth growth to support AI infrastructure demands. In fact, from 2020 to 2025, our WAN bandwidth grew a whopping 7x.

Multi-shard isolation, region isolation, and protective reroute: Each of our network shards has its own control plane, data plane, and management plane, and operates independently of other shards. This multi-shard isolation enables a high level of resiliency that’s rare for global backbones at our scale; in fact, it parallels the level of resiliency typically achieved via multiple independent global ISPs, without the associated complexity of managing multiple networks. Regional isolation minimizes the impact of failures and limits the impact radius. Protective ReRoute, a transport technique for shortening user-visible outages, glues it all together – it lets hosts promptly detect and route around any network failures within a few seconds. With Protective ReRoute deployed in our network, we have observed up to 93% reduction in cumulative outage minutes. 

Fully intent-driven, fine-grained programmability: We’ve built a highly programmable network with SDN controllers, standard APIs, and universal network models such as the Multi-Abstraction-Layer Topology representation, or MALT. This enables fully intent-driven network controls that allow us to tailor our network to specific application needs, and meet the unique needs of our customers. For example, these controls can be used for regulatory compliance and data sovereignty, including control over data in motion.  

Autonomous network: Over the last decade, we’ve transformed our network, moving from event-driven to machine-driven to now autonomous operations. This journey is fueled by ML, which provides us with actionable intelligence. Inspired by Google DeepMind’s work with graph neural networks (GNN) for accurate arrival-time predictions in Google Maps, we used GNN to create a digital twin of our network. This twin lets us predict and prevent outages, quickly pinpoint failures and their root causes, and optimize network capacity planning. As a result, we’ve observed failure mitigation times improve from hours to minutes, boosting our network’s efficiency and resilience with minimal human intervention. 

A network to unlock the full potential of AI

For cloud customers, Google’s global network offers the capacity, elasticity, and scale to deploy and leverage AI effectively, 24×7 app resilience with a reliable network, security through zero-trust principles, and performance that meets the needs of AI/ML applications. Furthermore, AI-driven efficiencies reduce maintenance toil, enable faster recovery, and improve ROI. And with Cloud WAN, starting today, Google Cloud customers can use Google’s global network to connect their global enterprises. For end users, this translates to expanded global reach, resilient mission-critical applications, zero-trust security to protect their data, and a performant network for power-intensive real-time apps. Taken together, these help ensure a great user experience. 

This is a truly exciting time as we continue to push the boundaries of network technology and realize the transformative potential it holds for our customers in the AI era. 

To learn more, we invite you to join us at our Google Cloud Next 2025 session, where we’ll share more details and demonstrate how our network continues to uphold Google’s mission and drive our customers’ success in the Gemini era. Keep an eye out for future blogs about the groundbreaking innovations that are powering Google’s next-generation global network.

Like many of you reading this, I fancy myself a builder. I got my first taste of making a computer do something based on code I wrote into a TI-99  console when I was 8 years old and was hooked. In the four decades that have followed, I’ve seen hundreds, maybe even thousands of developers and companies transform as we shifted from the internet to client-server to mobile and to cloud. 

I’ve not seen anything quite as transformative quite as fast as what we’re seeing now, helping companies harness AI to transform their software development, business process, information retrieval, and more. 

As part of our commitment to offering the most open and innovative generative AI partner ecosystem, we’re lucky enough to work with the most cutting-edge software companies in the world. As we head into Google Cloud Next ‘25 this week, we thought it might be inspiring to showcase the following 34 use cases illustrating how companies are using AI to transform how they work — moving beyond hype and into production. And if you want even more examples, check out our list of hundreds of real-world gen AI use cases we’ve helped build with our customers and partners.

Hope to see you at Next!

• AES – Enabling a massive audit process overhaul with accuracy, speed, and scale with gen AI agents built using Vertex AI and Anthropic’s Claude models.

• Airwallex – Detecting and managing fraud in real time in a scalable, always-available environment with Vertex AI, GKE, and GitLab. 

• Allegro – Enabling millions of real-time, personalized conversations with AI-powered omnichannel orchestration with Google Cloud and GrowthLoop.

• Augment Code – Accelerating AI coding with Google Cloud and NVIDIA and deepening the understanding of enterprise code with Vertex AI and Anthropic’s Claude.

• BrainLogic – Improving digital commerce experiences with Zapia, a uniquely Latin American personal AI assistant, built with Vertex AI and Anthropic’s Claude models.

• Bud Financial – Creating new ways of working and new operating models for financial institutions with Vertex AI and Astra DB from DataStax.

• Capital Energy – Taking sustainable energy to new heights with secure cloud technologies, AI, and data platforms with Vertex AI and Fortinet.

• CERC – Processing millions of credit data forecasts in minutes with near real-time insights and reduced costs with Databricks on Google Cloud.

• Copel – Providing real-time insights with a natural language AI agent that interacts directly with SAP ERP using Google Cloud Cortex Framework.

• Cox 2M – Reducing time to insights for non-technical users by 88% with natural-language chat and gen AI with Gemini and ThoughtSpot.

• Far Eastern New Century – Streamlining cross-border operations with generative AI with Google Cloud VMware Engine and Microfusion.

• Flockx – Combating loneliness by connecting individuals with AI Agent communities via their Collaboration Layer built with Google Cloud and Elastic. 

• HDFC ERGO – Pioneering insurance superapps for India with personalized AI-driven services at scale with Vertex AI, Niveus Solutions, and Lumiq.

• Hunkemöller – Building advanced AI solutions and a game-changing customer data platform for an SAP shop with Vertex AI and Devoteam G Cloud.

• Israel Antiquities Authority – Preserving the past for future generations with modern research models with Gemini, NetApp, VMWare, and CommIT.

• LiveX AI – Reducing customer support costs by up to 85% with AI agents trained and served on Google Kubernetes Engine and NVIDIA AI.

• L’Oréal – Delivering a flexible and scalable set of declarative generative AI APIs,  in 3 months, for secure access to gen AI for all with Gemini and LangChain.

• Maqqie – Transforming the future of HR recruitment and driving increased revenue and retention with Vertex AI and Rappit.

• Mercari – Dramatically improving the user experience for both buyers and sellers on its ecommerce marketplace with Vertex AI and Weights & Biases.

• Naologic – Democratizing access to AI for all companies—regardless of IT expertise—with Gemini, Google Cloud for Startups Cloud Program, and MongoDB.

• NotCo – Transforming its operations with Google Cloud AI agents and Google Cloud Cortex Framework with SAP and Eleven Solutions. 

• Palo Alto Networks – Speeding software development with Vertex AI and Anthropic’s Claude and protecting cybersecurity with NVIDIA and Google Cloud.

• Quora – Powering millions of human-like daily interactions through dynamic gen AI chat experiences with Gemini, Vertex AI, and Anthropic’s Claude.

• Rapid7 – Improving customer experiences with efficient cybersecurity customer support, at lightning speed, with Gemini and Ask AI.

• Replit – Making software creation accessible and efficient for everyone with Gemini and Anthropic’s Claude.

• Stax AI – Automating manual processes and transforming massive volumes of trust accounting data in minutes with Google Cloud and MongoDB.

• Studiosus Reisen – Embracing the future of travel with real-time customer bookings with Vertex AI, SAP, happtiq, and Solid Cloud.

• Sumitomo Rubber Industries – Accelerating software development time and delivering innovative products faster with Gemini and Kyocera.

• Suzano – Enhancing cloud-based data sources for sustainable materials management on SAP with Gemini, Google Cloud Cortex Framework, and Sauter.

• UDN Group – Improving clickthrough rates, enhancing operational efficiency, and providing highly reliable digital media services with Vertex AI and Merkle.

• Vodafone – Providing clear customer insights and actionable decision intelligence with a 360-customer view for improved service experiences with Google Cloud and Quantexa.

• Wayfair – Driving AI innovation in e-commerce with curated options that elevate online shopping experiences using Google Cloud and Snorkel

• WealthAPI – Delivering next-gen financial insights in real time to millions of customers for personalized guidance at scale with Gemini and DataStax Astra DB.

• Writer – Building and training over 17 large language models (LLMs) that scale up to 70 billion parameters for custom AI models with Google Cloud and NVIDIA.

Learn more at Google Cloud Next ‘25.

These partners are driving a wave of generative AI innovation, building next-generation applications and models, and integrating their solutions with powerful Gemini, Vertex AI, and BigQuery features. These advancements are helping customers realize the value of generative AI faster than ever before. Explore more stories to learn how our open partner ecosystem is supporting customer gen AI deployments, and discover how our partners simplify the discovery, procurement, and deployment of groundbreaking generative AI solutions in the Google Cloud Marketplace.

Every enterprise will soon rely on multi-agent systems – multiple AI agents working together – even when built on different frameworks or providers. Agents are intelligent systems that can act on your behalf using reasoning, planning, and memory capabilities. Under your supervision, they can think multiple steps ahead and accomplish tasks across various systems.

Multi-agent systems rely on models with enhanced reasoning capabilities, like those available in Gemini 2.5. They also depend on integration with your workflows and connection to your enterprise data. Vertex AI – our comprehensive platform to orchestrate the three pillars of production AI: models, data, and agents – seamlessly brings these elements together. It uniquely combines an open approach with comprehensive platform capabilities to ensure agents perform reliably: a combination that would otherwise require fragmented and fragile solutions.

Today, we’re announcing multiple enhancements to Vertex AI so you can: 

• Build agents with an open approach and deploy them with enterprise-grade controls

• Agent Development Kit (ADK) is an open-source framework for designing agents built on the same framework that powers Google Agentspace and Google Customer Engagement Suite (CES) agents. Many powerful examples and extensible sample agents are readily available in Agent Garden.

• Agent Engine is a fully managed runtime in Vertex AI that helps you deploy your custom agents to production with built-in testing, release, and reliability at a global, secure scale. 

• Agent2Agent protocol gives your agents a common, open language to collaborate – no matter which framework or vendor they are built on. We are driving this open initiative, partnering with 50+ industry leaders (and growing) to advance our shared vision of multi-agent systems.

• Equip agents with your data using open standards like Model Context Protocol (MCP) or connect directly with APIs and connectors managed in Google Cloud. You can ground your AI responses in Google Search, your preferred data sources, or with Google Maps data.

Introducing Agent Development Kit and Agent Garden: Building agents with an open approach

Agent Development Kit (ADK) is our new open-source framework that simplifies the process of building agents and sophisticated multi-agent systems while maintaining precise control over agent behavior. With ADK, you can build an AI agent in under 100 lines of intuitive code. Check out the examples here.

Currently available in Python (more languages coming later in the year), you can:

• Shape how your agents think, reason, and collaborate through deterministic guardrails and orchestration controls, giving you precise control over agent behavior and decision-making processes. 
• Interact with your agents in human-like conversations with ADK’s unique bidirectional audio and video streaming capabilities. With just a few lines of code, you can create natural interactions that change how you work with agents – moving beyond text into rich, interactive dialogue. Check out the demo of an interactive agent from the opening keynote at NEXT 2025 built on the ADK here.
• Jumpstart your development with Agent Garden, a collection of ready-to-use samples and tools directly accessible within ADK. Leverage pre-built agent patterns and components to accelerate your development process and learn from working examples.
• Choose the model that works best for your needs. ADK works with your model of choice – whether it is Gemini or your any model accessible via Model Garden. Beyond Google’s models, you can choose across 200+ models from providers like Anthropic, Meta, Mistral AI, AI21 Labs, CAMB.AI, Qodo, and more. 
• Select your deployment target, be it local debugging or any containerized production deployment, such as Cloud Run, Kubernetes, or Vertex AI. ADK also supports Model Context Protocol (MCP), enabling secure connections between your data and agents.
• Deploy to production using the direct integration to Vertex AI. This clear, reliable path from development to enterprise-grade deployment removes the typical overhead associated with moving agents to production.

While ADK works with your preferred tools, it’s optimized for Gemini and Vertex AI. For example, AI agents built with ADK using Gemini 2.5 Pro Experimental can break down complex problems through Gemini’s enhanced reasoning capabilities, and work with your preferred systems through its tool use capabilities. You can also deploy this agent to a fully-managed runtime and operate it at enterprise scale, using the native integration to Vertex AI from ADK.

Hear how our customers are already using ADK:

“Using Agent Development Kit, Revionics is building a multi-agent system to help retailers set prices based on their business logic — such as staying competitive while maintaining margins — and accurately forecasting the impact of price changes. ADK streamlines multi-agent transfer and planning, such as knowing when to transfer between specialized agents (data retrieval) and tools (constraint application), thereby combining Revionics’ pricing AI with agentic AI to automate entire pricing workflows. Data is central to Revionics’ process, and the development kit enables agents to efficiently reason over big data through storage artifacts rather than relying solely on the LLM context.” – Aakriti Bhargava, VP of Product Engineering and AI at Revionics.

“We used the ADK to develop an agent that ensures we’re installing EV chargers where drivers need them most. The agent assists our data analysts to leverage geographical, zoning, and traffic data to inform and prioritize critical EV infrastructure investments that maximize driver convenience with less strain on our teams.” – Laurent Giraud, Chief Data (&AI) Officer, Renault Group. 

“We’ve implemented the Agent Engine as the backbone of our video analysis AI agent, powered by Gemini. This setup allows us to leverage the Python Vertex AI SDK without worrying about infrastructure, saving us an estimated month of development time. Plus, the Agent Engine’s API seamlessly connects with other Google Cloud products like Workflows, giving us excellent maintainability and room to grow.” – Rina Tsuji, Senior Manager, Corporate Strategy, Nippon Television Holdings, Inc.

Introducing Agent Engine: Deploying AI agents with enterprise-grade controls

Agent Engine is our fully managed runtime that makes it easy to deploy AI agents to production. No more rebuilding your agent system when moving from prototype to production. Agent Engine handles agent context, infrastructure management, scaling complexities, security, evaluation, and monitoring. Agent Engine also integrates with ADK (or your preferred framework) for a frictionless develop-to-deploy experience. Together, you can:

• Deploy agents built using any framework – whether you’re using ADK, LangGraph, Crew.ai, or others, and regardless of your chosen model (Gemini, Anthropic’s Claude, Mistral AI, or others). This flexibility is paired with enterprise-grade controls for governance and compliance.
• Keep the context in your sessions: Rather than starting from a blank slate each time, the Agent Engine supports short-term memory and long-term memory. This way, you can manage your sessions and your agents can recall your past conversations and preferences.
• Measure and improve agent quality with comprehensive evaluation tools from Vertex AI. Improve agent performance by using the Example Store or fine-tune models to refine your agents based on real-world usage.
• Drive broader adoption by connecting to Agentspace: You can register your agents hosted on Agent Engine to Google Agentspace. This enterprise platform puts Gemini, Google-quality search, and powerful agents in the hands of employees while maintaining centralized governance and security.

Here’s how it all comes together:

In the coming months, we will further expand Agent Engine capabilities with advanced tooling and testing. Your agents will have computer-use capabilities and will be able to execute code. Additionally, a dedicated simulation environment will let you rigorously test agents with diverse user personas and realistic tools to ensure reliability in production.

Introducing Agent2Agent protocol: Connecting agents across your enterprise ecosystem

One of the biggest challenges in enterprise AI adoption is getting agents built on different frameworks and vendors to work together. That’s why we partnered with many industry leaders who share our vision of multi-agent systems to create an open Agent2Agent (A2A) protocol. 

Agent2Agent protocol enables agents across different ecosystems to communicate with each other,  irrespective of the framework (ADK, LangGraph, Crew.ai, or others) or vendor they are built on. Using A2A, agents can publish their capabilities and negotiate how they will interact with users (via text, forms, or bidirectional audio/video) – all while working securely together. 

As of today, 50+ partners such as Box, Deloitte, Elastic, PayPal, Salesforce, ServiceNow, UiPath, UKG, Weights & Biases, and many more are committed to working with us on the protocol. For details on the partners using the protocol, please refer to the blog here.

Beyond working with other agents, your agents also need access to your enterprise truth – the ecosystem of information you have built across data sources, APIs, and business capabilities. You can equip agents with your existing enterprise truth data without building from scratch, using any approach you prefer: 

• ADK supports Model Context Protocol (MCP), so your agents connect to the vast and diverse data sources or capabilities you already rely on by leveraging the growing ecosystem of MCP-compatible tools.

• From ADK, you can also connect your agents directly to your enterprise systems and capabilities. This includes 100+ pre-built connectors, workflows built with Application Integration, or data stored within your systems like AlloyDB, BigQuery, NetApp and much more. For example, you can build AI agents directly on your existing NetApp data, no data duplication required.

• Using ADK, you can also seamlessly connect to your existing agents built in other frameworks like LangGraph or call tools from diverse sources including MCP, LangChain, CrewAI, Application Integration, and any OpenAPI endpoints.

• In Apigee API management, we manage over 800K APIs that power your business, within and beyond Google Cloud. Using ADK, your agents can also tap into these existing API investments – no matter where they reside – with proper permissions.

Once connected, you can ground your AI responses with information like Google Search or specialized data from providers like Cotality, Dun & Bradstreet, HGInsights, S&P Global, and Zoominfo. For agents that rely on geospatial context, today we’re making it possible to ground your agents with Google Maps¹. We make 100 million updates to Maps data every day, ensuring it is fresh and factual. And now, using Grounding with Google Maps, your agents can provide responses with geospatial information tied to millions of places in the U.S. 

Enterprise grade security for your AI agents: Building agents your enterprise can trust

Beyond functionality, enterprise AI agents operating in production face security concerns like prompt injection attacks, unauthorized data access, and generating inappropriate content. Building with Gemini and Vertex AI in Google Cloud addresses these challenges in multiple layers. You can: 

• Control agent output using Gemini’s built-in safety features including configurable content filters and system instructions that define boundaries around prohibited topics and align with your brand voice.

• Manage agent permissions through identity controls that let you determine whether agents operate with dedicated service accounts or on behalf of individual users, preventing privilege escalation and unauthorized access.

• Protect sensitive data by confining agent activity within secure perimeters using Google Cloud’s VPC service controls, preventing data exfiltration and limiting potential impact radius.

• Establish guardrails around your agents to control interactions at every step – from screening inputs before they reach models to validating parameters before tool execution. You can configure defensive boundaries that enforce policies like restricting database queries to specific tables or adding safety validators using lightweight models.

• Auto-monitor agent behavior using comprehensive tracing capabilities that give you visibility into every action an agent takes, including its reasoning process, tool selection, and execution paths.

Get started building multi-agent systems 

The real value of Vertex AI isn’t just the individual capabilities outlined above, but in how they work together as an integrated whole. What previously required piecing together fragmented solutions from multiple vendors now flows seamlessly through a single platform. This unified approach eliminates painful tradeoffs between models, integration with enterprise apps and data, or production readiness. The result isn’t just faster development – it’s more reliable agents ready for enterprise workflows. To get started today: 

1. Build with Agent Development Kit
2. Visit the Vertex AI console
3. Explore our documentation

^{1. Grounding with Google Maps is currently available as an experimental release in the United States, providing access to only places data in the United States.}

Today’s fast-paced environment demands more than just data access; it requires a real-time data activation flywheel. A new reality is emerging where AI, infused directly into the data landscape, works hand-in-hand with intelligent agents. These agents act as catalysts, unlocking insights for everyone, and enabling the autonomous, real-time action that’s critical for success. Google’s Data & AI Cloud is built to power this flywheel, bringing AI to data for continuous, real-time data activation — a focus that’s attracting 5x more organizations to BigQuery than the two leading cloud companies that exclusively offer data warehouse and data science platforms.

Examples of leading companies include:

• Radisson Hotel Group boosted campaign productivity by 50% and revenue by over 20% using Gemini model fine-tuning with BigQuery. 

• Gordon Food Service used BigQuery to unify 170+ data sources, creating a scalable modern data architecture and an AI-ready foundation. This improved real-time response to critical business needs, enabled comprehensive analytics, driving significant customer adoption increase for their ordering applications, providing timely insights to their employees, while reducing costs and gaining market share. 

• J.B. Hunt is transforming logistics for shippers and carriers by consolidating fragmented systems, including Databricks, onto a unified BigQuery platform. 

• General Mills saves over $100 million using BigQuery and Vertex AI by giving employees secure access to LLMs to answer questions based on structured and unstructured data.

“We didn’t just need a place to store or consume data, we wanted a collaborator that could help us scale the most advanced data management in the industry.” – Jaime Montemayor, Chief Digital & Technology Officer, General Mills

Today, we’re announcing several new innovations with our autonomous data to AI platform powered by BigQuery, alongside our unified, trusted, and conversational BI platform with Looker:

1. Specialized agents for every user: New assistive and agentic experiences, grounded in your trusted data and available in BigQuery and Looker, are set to simplify and accelerate the work of data engineers, data scientists, analysts and business users.

2. Accelerating data science and advanced analytics: We are enhancing data science workflows in BigQuery with new AI-assisted notebooks and unlocking new insights with our BigQuery AI Query Engine, alongside seamless integration with real-time and open-source technologies.

3. Autonomous data foundation: New autonomous capabilities in BigQuery capture, manage, and orchestrate all data types, including native support for unstructured data handling and open data formats like Iceberg.

Let’s take a deeper look at each of these developments.

1. Specialized agents for every user

We believe AI should be accessible to everyone. We have made AI driven assistive experiences broadly available in BigQuery and Looker, and now we have expanded to specialized agents that best meet the needs for all data roles, including:

• Data engineering agent capabilities, embedded in BigQuery pipelines (GA), deliver support to build data pipelines, perform data preparation (GA) like transformation and enrichment of data, maintain data quality with anomaly detection (preview), and automate metadata generation. Traditionally data engineers spend countless hours cleaning, transforming, and validating data — these agents replace tedious and time consuming tasks and enable trusted data, boosting productivity of your data teams.

• Data science agent (GA), embedded within Google’s Colab notebook, enables every stage of model development. It automates feature engineering, provides intelligent model selection, enables scalable training, and faster iteration. This agent allows data science teams to focus on building advanced data science workflows, instead of wrestling with data and infrastructure.

• Looker conversational analytics (preview) empowers every user to interact with data using natural language. Expanded capabilities developed in partnership with DeepMind, not only conducts advanced analysis but also explains its thinking transparently, empowering all users to understand the agent’s behavior and seamlessly resolve ambiguities. In addition, Looker’s semantic layer improves accuracy by as much as two thirds. As users reference business terms like ‘revenue’ or ‘segments,’ the agent knows exactly what you mean and can calculate metrics in real-time, ensuring it delivers accurate, relevant, and trusted results. Additionally, we are launching a conversational analytics API (preview) for developers to build and embed conversational analytics into applications and workflow.

To power intelligence across assistive and agentic experiences in the BigQuery autonomous data to AI platform, we are also launching BigQuery knowledge engine (preview). It leverages the power of Gemini to analyze schema relationships, table descriptions, and query histories to generate metadata on the fly, model data relationships, and recommend business glossary terms. This knowledge engine is the foundation of AI powered experiences, including AI-powered data insights and semantic search (GA) across BigQuery, grounding AI and agents in business context.

Today all of our Gemini powered assistive and agentic experiences in BigQuery and Looker are available to all customers within the existing tiers of our pricing models — no add-ons!

2. Accelerating data science and advanced analytics

Our BigQuery autonomous data to AI platform is fundamentally changing how data scientists and analysts work, enabling new AI driven data science experiences and new engines to process complex data and support advanced analyses in real-time.

First, we’re supercharging the BigQuery notebook experience with AI. We are introducing intelligent SQL cells that understand your data’s context and provide smart suggestions as you write code and enable you to join data sources directly within your notebook. We are also adding native exploratory analysis and visualization capabilities, making it easy to explore data, as well as adding features to enable easier collaboration with colleagues. Data scientists can also schedule analyses to run and refresh insights periodically. And, to share insights more broadly across the organization, we are introducing the ability to build interactive data apps – dynamic, user-friendly interfaces powered by your notebook.

Building on this enhanced notebook environment, we are also announcing BigQuery AI query engine to support advanced, AI-driven analytics. This engine enables data scientists to move beyond simply retrieving structured data to seamlessly processing both structured and unstructured data together with added real-world context. The BigQuery AI query engine co-processes traditional SQL alongside Gemini to inject runtime access to real-world knowledge, linguistic understanding, and reasoning abilities. A data scientist can now ask questions like: “Which products in our inventory are primarily manufactured in countries with emerging economies?” The foundation model inherently knows which countries are considered emerging economies. Analysts can ask: “Which products are included in these social media images?” and our new engine processes the unstructured images and matches them to your product catalog. This engine supports a broad range of use cases, including building richer features for models, performing nuanced segmentation, and uncovering insights previously out of reach.

Furthermore, we empower users with the best of the open-source ecosystem, enhanced for the cloud. Google Cloud for Apache Kafka (GA) facilitates real-time data pipelines for event sourcing, model scoring, messaging and real-time analytics, powering serverless execution of Apache Spark workloads within BigQuery (preview). Customer use of our serverless Spark capability has nearly doubled in the past year, and we have enhanced this engine to provide 2.7x faster processing than the prior year.

BigQuery allows data scientists to leverage the tools they need on Google’s serverless and scalable architecture, whether it’s SQL, Spark, or the semantic power of foundation models, enabling faster innovation without traditional infrastructure challenges.

“We see SQL and Spark as two complementary ways of accessing and transforming data. Spark is especially useful to us in use cases that require complex business logic, which although niche, are extremely business-critical. Having an unified platform for SQL, Spark and AI, with the development experience in notebooks will considerably simplify these critical use cases.” – Andrés Sopeña Pérez, Head of Data and AI, Trivago 

3. An autonomous data foundation across the entire data lifecycle 

Underpinning our specialized agents and advanced analytics engines is an autonomous data foundation designed for the complexity of modern data. We are fundamentally changing the landscape by making unstructured data a first-class citizen within BigQuery. This is achieved with new capabilities in the platform including autonomous and invisible governance, orchestration for diverse data workloads, and a commitment to flexibility via open formats, ensuring your data is always ready for any data science or AI challenge. And we do all of this while minimizing operational overhead and delivering optimal price performance.

The biggest untapped opportunity for many organizations lies in the potential of their unstructured data. While structured data has pathways to analysis, unique insights embedded in images, audio, video, and text are often hard to extract and underutilized, and typically reside in siloed systems. BigQuery directly tackles this challenge by making unstructured data a first-class citizen with multimodal tables (preview), allowing you to bring rich, complex data types alongside structured data for unified storage and querying. To effectively manage this comprehensive data estate, our enhanced BigQuery governance (preview) provides a single, unified view for data stewards and professionals to handle discovery, classification, curation, quality, usage, and sharing, including automated cataloging (GA) and metadata generation (experimental). Moreover, to ensure timely insights from all your data streams, BigQuery continuous queries (GA) enable instant analysis and action on streaming data using SQL, regardless of its original format.

Our advanced support of multimodal data, both structured and unstructured, is driving adoption – customer use of Google’s AI models in BigQuery for multimodal analysis has grown by 16x year over year. Our integrated approach to data and AI is also cost effective, together BigQuery and Vertex AI are between 8-16x more cost efficient when compared to other independent data warehouse and AI platforms.

Our commitment to an open ecosystem remains paramount. BigQuery tables for Apache Iceberg (preview), delivers the flexibility of an open data lakehouse alongside the performance and integrated tooling of BigQuery, allowing you to connect your Iceberg data to SQL, Spark, AI and third party engines in an open and interoperable manner. This offering provides adaptive and autonomous table management, delivers high-performance streaming, auto-AI generated insights, near infinite serverless scale, and advanced governance. Through integration with Cloud storage, our managed service provides centralized fine-grained access control management and fail-safe capabilities.

Finally, the autonomous data to AI platform is self-optimizing. It scales resources, manages workloads, and helps ensure their cost-effectiveness with advanced workload management capabilities (GA). Furthermore, we’ve simplified purchasing with the new BigQuery spend commit (GA), unifying spend across our BigQuery platform, providing flexibility to move spend across data processing engines, streaming, governance, and more.

Get started on your data and AI journey by taking advantage of our BigQuery data migration offer. We can’t wait to learn about the ways you are innovating using data.

Millions of developers use Firebase to engage their users, powering over 70 billion instances of apps every day, everywhere — from mobile devices and web browsers, to embedded platforms and agentic experiences. But full-stack development is evolving quickly, and the rise of generative AI has transformed not only how apps are built, but also what types of apps are possible. This drives greater complexity, and puts developers under immense pressure to keep up with many new technologies that they need to manually stitch together. Meanwhile, businesses of all sizes are seeking ways to make AI app development cycles more efficient, deliver quality software, and get to market faster. 

Today at Google Cloud Next, we’re introducing a suite of new capabilities that transforms Firebase into an end-to-end platform to accelerate the complete application lifecycle. The new Firebase Studio, available to everyone in preview, is a cloud-based, agentic development environment powered by Gemini that includes everything developers need to create and publish production-quality AI apps quickly, all in one place. Several more updates across the Firebase platform are helping developers unleash their modern, data-driven apps on Google Cloud. These announcements will empower developers to forge new paths for building AI applications across multiple platforms. 

Meet Firebase Studio

Over the past year, we launched many new services, including Gemini in Firebase, Genkit, and Project IDX (a fork of Code OSS), to make building AI apps faster and easier. We’re taking a significant step forward with the launch of Firebase Studio, which fuses all of these capabilities together with Firebase services and the creative power of Gemini into a new, natively agentic experience.

For new apps, choose from over 60 pre-built templates or get started with the App Prototyping agent. It assists you with designing your app — such as the UI, API schema, and AI flows — all using natural language, images, drawing tools, and screenshots. Keep prompting to iterate on your prototype, and when ready, deploy it directly to Firebase App Hosting. Share a URL to a fully functional version of your prototype to get feedback or run experiments. Monitor the usage and behavior at a glance or jump into the Firebase Console for more detailed monitoring. At any time, you can open your app inside a Firebase Studio coding workspace with a single click and no additional setup. There, you can refine the architecture and expand features to prepare for production deployment.

Coding workspaces also enable you to: 

• Simplify coding workflows: Write code and test features, all with assistance from Gemini in Firebase at every step of the way. Complete a variety of tasks like debugging, testing, refactoring, explaining, and documenting code with ease.

• Enhance existing apps: Import existing codebases from your local machine or git-based repositories, including GitHub, GitLab, and Bitbucket. Create custom templates for your preferred tech stacks to share across your team. 

• Create full-stack experiences: Customize and evolve all aspects of your apps, from AI model inference, agents, and retrieval augmented generation (RAG), to the user experience, business logic, database, and more. Easily expose and integrate tools such as APIs and microservices to your AI apps.

• Work with familiar tools: Bring along your specific configurations, such as system tools, extensions, and environment variables to tailor your workspaces. Access thousands of extensions from the Open VSX Registry.

• Flexible deployment options: Set up your app to run on the cloud with built-in integrations to Firebase backend services and Google Cloud Run. You can also deploy on your own custom infrastructure.

Firebase Studio is currently available with three workspaces at no cost during preview. Members of the Google Developer Program get up to 30 workspaces. Check out Firebase Studio today.

Engage AI agents throughout your entire workflow

We’re also providing early access to Gemini Code Assist agents from within Firebase Studio. For example, you might invoke the Migration agent in Firebase Studio to help you migrate your code between versions of Java; the AI Testing agent to run adversarial tests against AI models to uncover and fix potentially harmful outputs; and the Code Documentation agent to chat with a wiki-style knowledge base about your code to ease the onboarding of new team members. 

Join the Gemini Code Assist agents wait list today via the Google Developer Program.

Firebase App Distribution is a unified mobile app testing service for running manual and automated tests. The new App Testing agent in Firebase App Distribution can simulate real-world user interactions with your app. For example, you can write a test that sets a goal to “Find a trip to Greece.” The App Testing Agent will use Gemini to formulate a plan to achieve that goal, and run it on virtual or physical devices, navigating your UI and producing detailed pass/fail results with intuitive rationales and visuals of the paths that the agent chose. The App Testing agent is available now in preview and you can try it out on your Android app today, with more platforms coming this year.

Create new AI app experiences

A lack of best practices and standards create challenges when integrating cutting-edge AI features into your applications. That’s why we’re continuing to invest in robust frameworks, SDKs, and tooling to help streamline the development process, freeing you to focus on crafting truly engaging and innovative user interactions.

Expanded language support for Genkit
Genkit helps reduce the complexity of building, testing, and monitoring your apps’ AI features. Develop powerful agentic experiences with support for structured output, tool calling, human-in-the-loop interactions, retrieval augmented generation (RAG), Model Context Protocol (MCP), and multi-model orchestration. Today, we’re making it easier to do that in your preferred language, by introducing early support for Python and expanded support for Go. Access Gemini models, Imagen 3, and additional models such as Llama and Mistral through Vertex Model Garden, plus self-hosted models with Ollama, and a growing ecosystem of third-party models using community plugins. 

Try this template in Firebase Studio to build with Genkit.

New models through Vertex AI in Firebase
Vertex AI in Firebase lets developers integrate generative AI into their applications by providing a streamlined, secure SDK. It’s used by thousands of apps today, such as Meal Planner, a meal planning and shopping list management app; Life, an AI-powered diary assistant; HiiKER, an offline hiking maps provider; and Waveful, a social media app for creators. In March, we added support for Imagen 3 models (Imagen 3 and Imagen 3 Fast), in addition to the Gemini family of models, which lets you add image generation directly to your Android, iOS, Flutter, and Web applications. Today, we are adding support for the Gemini 2.0 Multimodal Live API, enabling more conversational interactions in apps, like allowing customers to ask audio questions and get responses.

Accelerate modern, data-driven apps 

We’re also providing you with greater control of your app architecture and deployment processes with Firebase Data Connect and Firebase App Hosting, now generally available. 

Build sophisticated apps with Firebase Data Connect
Firebase Data Connect offers the robust reliability of Google Cloud SQL for PostgreSQL with instant GraphQL APIs and type-safe SDKs. Build a wide range of experiences, like social media apps with complex user relationships, e-commerce platforms with large product catalogs, or personalized recommendations with built-in vector search.

Data Connect now helps you to:

• Easily generate schemas and queries: Use Gemini in Firebase to automatically generate your Data Connect schemas, queries, mutations, and client SDKs, significantly speeding up backend development. 

• Leverage expanded query capabilities: Data Connect now has expanded query power with native aggregation support for deeper data insights, atomic data modifications, and transactions with server value expressions, helping to ensure data integrity across complex operations.

• Build with web frameworks: Enjoy tight integration and streamlined data handling with generated type-safe hooks and components for web frameworks, enabling rapid development of dynamic, data-driven applications.

Deploy with Firebase App Hosting
Firebase App Hosting is an opinionated, git-centric hosting solution for modern, full-stack web apps. App Hosting accelerates time-to-market by managing your app’s entire stack, from the build, to the CDN, to server-side rendering. You push to GitHub, and App Hosting figures out the rest. App Hosting is built on enterprise-grade Google Cloud services: Cloud Build, Cloud Run, Cloud CDN, and more.

With this release of App Hosting, you can:

• Easily test and troubleshoot builds: App Hosting now has a  local emulator, and improved error messages to help you get ahead of and troubleshoot build failures.

• Recover from production incidents in seconds: Use App Hosting’s new monitoring dashboard to understand your app’s performance and health, and instantly roll back to a previous version if you spot a regression.

• Connect to a Virtual Private Cloud (VPC): Give your app access to backend services in your Google Cloud project that are not accessible on a public IP address (e.g., caching content with Cloud Memorystore, or accessing data from non-Firebase databases).

Get ready to reimagine not just how you build your apps, but also what kinds of apps you can build! Learn more about these products and more on the Firebase blog. We can’t wait to see what you create with the Firebase platform.

Today, we’re continuing to invest in generative media by adding Lyria, Google’s text-to-music model, to Vertex AI in preview with allowlist. With the addition of music, Vertex AI is now the only platform with generative media models across all modalities – video, image, speech, and music. This means you can build a complete, production ready asset starting from a text prompt, to an image, to a complete video asset with music and speech. 

In addition to Lyria, we’re launching new features and updates to improve our other generative media models:

1. New editing and camera control features for Veo 2, our advanced video generation model, are available in preview with allowlist to help our customers refine and repurpose video content with precision. This gives you creative control over your video, helping your teams iterate faster, produce higher-quality content, and reduce post-production time and costs. 

2. Chirp 3, our groundbreaking audio generation and understanding model, now includes Instant Custom Voice, a new way to create custom voices with just 10 seconds of audio input. You can also weave AI-powered narration into your existing recordings, and add a speech transcription capability that can distinguish between speakers. Both features are available through a preview with allowlist.  

3. Imagen 3, our highest quality text-to-image model, now has improved image generation and inpainting capabilities for reconstructing missing or damaged portions of an image. Our latest update significantly elevates the quality of object removal, delivering a more natural and seamless editing experience. 

In alignment with our AI Principles, the development and deployment of Lyria, Veo 2, Chirp 3, and Imagen 3 on Vertex AI prioritizes safety and responsibility with built-in precautions like digital watermarking via SynthID, safety filters, and data governance. And, with our industry-first approach to indemnification, you can use content generated with a range of our products knowing Google will indemnify you for third-party IP claims, including copyright.

Lyria: Text-to-music model now available on Vertex AI

Lyria produces high-fidelity audio, meticulously capturing subtle nuances and delivering rich, detailed compositions across a range of musical genres. Lyria on Vertex AI can help enterprises:

• Elevate brand experiences: Quickly create soundtracks for marketing campaigns, product launches, or immersive in-store experiences, all tailored to your brand’s unique identity. Lyria enables you to create sonic branding that resonates deeply with your target audience, fostering emotional connections and enhancing brand recall.

• Streamline content creation: For video production, podcasting, and digital content creation, finding the perfect royalty-free music can be a time-consuming and costly process. Lyria eliminates these hurdles, allowing you to generate custom music tracks in minutes, directly aligning with your content’s mood, pacing, and narrative. This can help accelerate production workflows and reduce licensing costs.

For example:

Craft a high-octane bebop tune. Prioritize dizzying saxophone and trumpet solos, trading complex phrases at lightning speed. The piano should provide percussive, chordal accompaniment, with walking bass and rapid-fire drums driving the frenetic energy. The tone should be exhilarating, and intense. Capture the feeling of a late-night, smoky jazz club, showcasing virtuosity and improvisation. The listener should not be able to sit still.

Expanding Veo 2 with a new robust set of editing features

Today, we’re announcing the preview of a robust feature set that helps you create videos, edit them, and add visual effects with Veo 2. These features help teams edit and repurpose video content to meet your evolving needs, transforming Veo on Vertex AI from a generation tool to a comprehensive video creation and editing platform. Now you can: 

• Refine and enhance existing footage with:

• Inpainting: Get clean, professional edits without manual retouching. You can remove unwanted background images, logos, or distractions from your videos, making them disappear smoothly and perfectly in every single frame, so it looks like they were never there.

• • Outpainting: Extend the frame of existing video footage, transforming traditional video into optimized formats for web and mobile platforms. This helps make it easy to adapt your content for various screen sizes and aspect ratios – for example, converting landscape video to portrait for social media shorts.

• Implement sophisticated cinematic techniques: New features include directing shot composition, camera angles, and pacing that help teams use sophisticated cinematic techniques with ease, without requiring complex prompting or specialized expertise. For example, you can use camera pre-sets to move the camera in different directions, create a timelapse effect, or generate a drone style shot.     

• Create a cohesive video by connecting two existing assets (interpolation): With interpolation, you can define the beginning and end of a video sequence, allowing Veo to seamlessly generate the connecting frames. This ensures smooth transitions and maintains visual continuity, creating a polished and professional final product.

Chirp 3: Instant Custom Voice and Transcription updates

Last month, we integrated Chirp 3, our groundbreaking audio understanding and generation model, into Vertex AI. Chirp 3’s new HD voices feature offers natural and realistic speech in over 35 languages with eight speaker options. 

Now, we’re announcing two new features: 

• Chirp 3: Instant Custom Voice is now generally available through an allowlist. Now, you can generate realistic custom voices from 10 seconds of audio input. This enables enterprises to personalize call centers, develop accessible content, and establish unique brand voices—all while maintaining a consistent brand identity. To ensure responsible use, Instant Custom Voice includes built-in safety features, and our allowlisting process involves rigorous diligence to verify proper voice usage permissions. 

• Chirp 3: Transcription with Diarization is now available in preview with allowlist. This powerful feature accurately separates and identifies individual speakers in multi-speaker recordings, significantly improving the clarity and usability of transcriptions for applications like meeting summaries, podcast analysis, and multi-party call recordings.

Imagen 3: Improvements to Imagen quality and editing

Over the last year we’ve made huge improvements to Imagen 3, our highest quality text-to-image model, capable of generating images with even better detail, richer lighting and fewer distracting artifacts than our previous models.

Imagen 3 Editing provides a powerful and user-friendly way to refine and tailor any image. We’ve made significant improvements to Imagen 3 inpainting capabilities for reconstructing missing or damaged portions of an image. Our latest update significantly elevates the quality of object removal, delivering a more natural and seamless editing experience. Here is an example of how you can quickly remove unwanted objects, blemishes, or distractions from your photos.

Build with enterprise safety and security 

Designing and developing AI to be secure, safe, and responsible is paramount. Consistent with our AI Principles, Lyria, Veo 2, Chirp 3, and Imagen 3 on Vertex AI were built with safety at the core. 

• Digital watermarking: Google DeepMind’s SynthID embeds invisible watermarks into every image, video and audio frame that Imagen, Veo, and Lyria produce, helping decrease misinformation and misattribution concerns.

• Safety filters: Veo, Imagen, Lyria, and Chirp all have built-in safeguards to help protect against the creation of harmful content and adhere to Google’s Responsible AI Principles. We will continue investing in new techniques to improve the safety and privacy protections of our models. 

• Data governance: We do not use customer data to train our models, in accordance with Google Cloud’s built-in data governance and privacy controls. Your customer data is only processed according to your instructions.

• Copyright indemnity: Our indemnity for covered generative AI services offers peace of mind for copyright concerns.

Customers are delivering value with generative media models on Vertex AI

Generative AI is no longer a futuristic concept, but a powerful tool driving real-world business results. Companies like WPP, Agoda, Bending Spoons, Monks.Flow, The Brandtech Group, and Bloomberg Connects are using our generative media models in production. Let’s look at some concrete examples of how leading enterprises are leveraging Google Cloud’s generative media capabilities:

• Goodby, Silverstein & Partners: In 1937, Salvador Dalí imagined “Giraffes on Horseback Salad” — a cinematic vision so surreal, so ahead of its time, that it proved impossible to produce. For almost a century, it lived only in sketches and notes. Now, with the power of Veo 2, Goodby Silverstein & Partners and The Dalí Museum have realized that vision — using tools finally capable of transforming surrealism into film.

  “Dalí imagined a film so surreal, so untethered from convention, that it couldn’t exist in his lifetime. Now, thanks to the astonishing capabilities of Veo 2 and Imagen 3, we’ve been able to help bring that vision to life—not as a replica, but as a reawakening. It’s one of the most creatively thrilling things we’ve ever done.” – Jeff Goodby, Co-Chairman, Goodby Silverstein & Partners.

L’Oreal Groupe:

L’Oreal Groupe is leveraging Veo and Imagen to transform the end-to-end production of high-quality video and image assets, helping foster greater creative exploration across their global marketing initiatives and upholding their commitment to trustworthy AI.

“By integrating Veo and Imagen into our creative process, we’re not just speeding up marketing content creation, we’re changing how we approach creativity. These models act as powerful creative partners, empowering our teams to experiment with new ideas and respond to the market. We’re expanding our qualitative  video and image production across 20 additional countries and languages, all while upholding our trustworthy AI values.” – Thomas Ménard, Manager of AI Center Enablement, L’Oreal Groupe

Kraft Heinz: 

Kraft Heinz’s Tastemaker platform empowers their teams with access to Veo 2 and Imagen 3, dramatically accelerating creative and campaign development processes. 

“With Veo 2 on Vertex AI as part of our Tastemaker platform, Kraft Heinz has unlocked unprecedented speed and efficiency in our creative workflows. What once took us eight weeks is now only taking eight hours, resulting in substantial cost savings. Implementing Google Cloud AI within our platform that is deeply trained on our brand intelligence, allows innovation and creative teams to rapidly prototype, test, and deploy content, transforming how we bring our iconic brands to life.” – Justin Thomas, Head Digital Experience & Growth

By leveraging our cutting-edge AI models on Vertex AI, enterprises are achieving remarkable gains in efficiency, creativity, and customer engagement. This momentum is a testament to the power of our technology and its ability to help drive tangible business value.

Get started

Get started with Veo, Imagen, and Chirp on Vertex AI today. To get started with Lyria, reach out to your Google Cloud account representative.

The most exciting startups in the world are in Las Vegas this week, as Google Cloud Next kicks off with a major focus on how AI and cloud are powering the next great wave of innovation. Our AI-optimized infrastructure, our broad AI model offerings, and our deep relationships with founders around the globe have put Google Cloud at the center of today’s AI startup movement — in fact, over the past two years, more than 60% of the world’s generative AI startups have been building on Google Cloud.

AI research labs and model builders like Anthropic, Cohere, Magic, Mistral, and AI21 Labs are great examples. Anthropic has been using Google Cloud infrastructure to support model training and inference for several years. Google Cloud has also become an increasingly important route for organizations to access Anthropic’s models. In fact, more than 4,000 companies have started using Claude models via Vertex AI thus far. 

This week at Google Cloud Next, we’re highlighting the progress that even more global startups are making toward building AI systems and applications that will create real value for people and businesses. We’ll share some of the new startups who have chosen Google Cloud as their key technology partner; new industry partnerships and resources to make it easier for early-stage startups to build and go to market on Google Cloud; new companies joining our global Accelerator programs; and more ways that startups are achieving success with Google Cloud.

Supporting more AI startups on Google Cloud

This week, we’re announcing an exciting group of startups who are launching new or expanded work with Google Cloud. These companies span a broad swath of use cases, products, and industries, including consumer apps used by millions of people, enterprise use cases like search and data analytics and storage, developer coding assistance, and solutions for unique vertical challenges. These include:

• Augment Code is running its AI coding assistant, which specializes in helping developers navigate and contribute to production-grade codebases, on Google Cloud and is using Anthropic’s Claude models through Vertex AI.
• Autoscience, a startup building AI agents to aid in scientific research, is using Google Cloud infrastructure and resources through our Startup Program as it begins to build and market its products.
• Anysphere, the startup behind the AI-powered code editor Cursor, is now using Anthropic’s Claude models on Google Cloud to scale its AI coding assistant to more developers.
• Big Sur AI now offers its AI-powered platform for retail and e-commerce customers on Google Cloud Marketplace.
• Captions recently released its integration with Veo 2, making it easy for users to add B-roll content to the startup’s talking videos. 
• Eon.io, a startup focused on enterprise backup and recovery, has begun working with Google Cloud through our partnership with Lightspeed Capital; it’s now adding new AI search capabilities to its platform.
• Features & Labels is a generative media platform for developers, accelerating the inference of gen AI models to improve the speed in which content is generated. The Fal team is working with Google Cloud to leverage its Veo 2 technology to help its users create videos with realistic motion and high-quality output. 
• Hebbia has integrated Gemini models into its Matrix platform, which helps organizations build AI agents capable of working across all of their data.
• Magic, which is building frontier-scale models capable of understanding very large codebases, is growing its use of GPUs on Google Cloud as it accelerates research and model training.
• Photoroom, a French startup that provides gen AI photo-editing and design capabilities to consumers and businesses, has used Veo 2 and Imagen 3 to improve the quality of its offering and accelerate its development. 
• Physical Intelligence recently partnered with Google Cloud to support model development, using our secure and scalable AI infrastructure.
• Spot AI is a video AI startup that transforms passive security cameras into AI Agents for improving security, safety, and operations in industries like manufacturing, retail, hospitals, construction and more. They’re using Google Cloud to power their new interface for building custom video AI agents, called Iris.
• Story is working with Google Cloud’s web3 services and infrastructure to bring new capabilities to developers on its platform.
• Studyhall AI, which graduated from our UK Growth Accelerator program, has built a mobile application that uses Gemini models to help coach students on reading, writing, and exam prep.
• Safe Superintelligence is partnering with Google Cloud to use TPUs to accelerate its research and development efforts toward building a safe, superintelligent AI.
• Synthesia, a startup that operates an AI video platform, is using Google Cloud to build the next generation of advanced AI models that replicate realistic human likenesses and voice; the startup is also using Gemini models to handle complex vision and language-based tasks with speed and accuracy.
• Ubie, a healthcare-focused startup founded in Japan, is using Gemini models via Google Cloud to power its physician assistance tool.
• Udio is using TPUs to help train its models for music generation and serve its rapidly growing customer base.
• Ufonia helps physicians deliver care by using AI to automate clinical consultations with patients. It is using Google Cloud’s full AI stack to power its platform, including infrastructure, models on Vertex AI Model Garden, BigQuery, and GKE.
• Wagestream, a financial services startup, is using Gemini models to handle more than 80% of its internal customer inquiries, including questions about pay dates, balances, and more.
• Wondercraft, an AI-powered content studio that helps users create engaging audio ads, podcasts and more, is leveraging Gemini models for some of its core functionalities and will soon release a Veo 2 integration.

New partnerships with accelerators and VCs

We’re also expanding our work with leading venture capital firms and accelerators, building on our existing relationships with firms like Sequoia and YCombinator. These partnerships help provide technology like TPUs and Gemini models to fast-growing startup companies who are building with AI.

Today, we’re announcing a significant new partnership with the leading venture capital firm Lightspeed, which will make it easier for Lightspeed-backed startups to access technology and resources through the Google for Startups Cloud Program. 

This includes upwards of $150,000 in cloud credits for Lightspeed’s AI portfolio companies, on top of existing credits available to all qualified startups through the Google for Startups Cloud Program. These credits help ensure participating startups will have more reliable access to cloud infrastructure and AI technology as they scale. 

Lightspeed portfolio companies have already been using Google Cloud infrastructure, AI, and data tools, including Augment, Contextual, Grafana, and Mistral.

New resources to help startups build and go to market more quickly

Today we’re announcing new resources through the Google for Startups Cloud Program that will help startups access our infrastructure and technology more quickly. 

Our Springboard program, which helps startup customers bring products to market on Google Cloud, has grown significantly. In the past year, startup customers have launched many new products on Google Cloud Marketplace through this program, including Ask-AI, Cascade AI, Cimulate, Ema.co, Fireworks AI, Mendel, Predibase, Resemble AI, Tektonic AI, and Vijil.

We are also announcing our Startup Perks program, which provides early stage startups with preferred access to solutions from our partners like Datadog, Elastic, ElevenLabs, GitLab, MongoDB, NVIDIA, Weights & Biases, and more. Exclusive discounts and benefits will be added on a regular basis, helping startups build and grow with the best of the Google Cloud ecosystem. 

Additionally, Google for Startups Cloud Program members will receive an additional $10,000 in credits to use exclusively on Partner Models through Vertex AI Model Garden, so they can quickly start using both Gemini models and models from partners like Anthropic and Meta.

We’re proud that 97% of companies who join our Google for Startups Cloud Program choose to stay with Google Cloud after their program credits expire, underscoring the value that our products are providing.

New accelerator cohort companies

Google Cloud currently offers a series of Accelerator programs for startups around the world. Today, we’re announcing the Spring 2025 cohort of companies in Google for Startups Cloud AI Accelerator for startups based in North America:

Future of AI: Perspectives for Startups

This year, we also launched our first-ever “Future of AI: Perspectives for Startups 2025” report. To gain a deeper understanding of where AI is headed, we gathered perspectives from 23 industry leaders and investors on their expectations for AI and what it means for startups in 2025 and beyond.

These experts weighed in on topics like the role of AI agents, the future of AI infrastructure, the areas startup investors are prioritizing, and much more. I encourage anyone involved in or interested in the AI startup world to take a look.

It’s this kind of fresh thinking and new technology that’s making Google Cloud a home for the startup community, and AI startups in particular — and we see Next ‘25 as an important extension of this growing ecosystem. With more that 60% of gen AI startups building on our platform today, we want to be  a key partner for their innovation. This commitment to startups is ongoing, and will continue to grow with new resources, accelerators, and other programs that help them build and scale their businesses. Please be sure to visit the Startup Hub on the Mandalay Bay Expo Floor at Next ‘25 to learn more.

Since the beginning, partners have been core to Google Cloud — and that’s been especially true in the AI era. I’m amazed at the ways they have helped bring both Google’s AI innovations and their own incredible AI products and services to customers. Partners have already built more than 1,000 AI agent use cases for customers across nearly every industry. 

The AI opportunity for Google Cloud partners is growing fast. For example, a new study from IDC¹ found that global systems integrators will grow their Google Cloud AI practices as much as 100% this year — and almost half of their Google Cloud AI projects have already moved into widespread production thanks to initial ROI that customers are seeing. It’s clear that much of the opportunity ahead lies in agentic AI — and now, partners are infused at every layer of our AI agent stack. 

This week at Google Cloud Next, we’re announcing updates that will help our partners build AI agents, power them with enterprise data and a choice of models, and bring them to customers, including through a new AI Agent Marketplace. We’re also launching a new open protocol, with support from more than 50 of the industry’s leading enterprise technology companies, which will allow AI agents to securely communicate in order to successfully complete tasks. And we’re enhancing the ways we go to market together, offering new incentives and resources for co-selling and training, as well as new AI offerings in Workspace for resellers.

We’re committed to helping all of our partners capitalize on the AI opportunity, whether they’re building new technology, integrating with our products, or delivering critical enterprise services.

Integrating our partners at every layer of the agentic AI stack

We’ve always taken an open approach to AI, and the same is true for agentic AI. With updates this week at Next ‘25, we’re now infusing partners at every layer of our agentic AI stack to enable multi-agent ecosystems. Here’s a closer look:

• Agent2Agent (A2A) protocol: Today, we’re launching a new open protocol, with support and contributions from our partners, that will allow AI agents to communicate with each other, securely exchange information, and coordinate actions across various enterprise platforms or services like Atlassian, Box, Cohere, Intuit, Langchain, MongoDB, PayPal, Salesforce, SAP, ServiceNow, UKG and Workday. We believe the A2A framework will add significant value for customers, whose AI agents will now be able to work across their entire enterprise application estates. More partners can begin building with the A2A framework today and can learn more in our technical blog here. 
• AI Agent Marketplace: We’re also launching a new AI Agent Marketplace — a dedicated section within Google Cloud Marketplace that will easily allow customers to browse, purchase, and manage AI agents built by our partners. Today, partners like Accenture, BigCommerce, Deloitte Elastic, UiPath, Typeface, and VMware are offering agents through our AI Agent Marketplace, with additional agents launching soon from partners like Cognizant, Slalom, and Wipro.
• Power agents with all your enterprise data: Data and AI models underpin all AI agents, and our open platform and agentic AI tooling make it possible for customers to train and fine-tune agents on their entire data estates. Today, we partner with companies like NetApp, Oracle, SAP, Salesforce, and ServiceNow — meaning customers with data stored in these popular platforms can now put this data to work to improve their AI agents.
• Expert AI services: Our ecosystem of services partners — including Accenture, BCG, Capgemini, Cognizant, Deloitte, HCLTech, Infosys, KPMG, McKinsey, PwC, TCS, and Wipro — have actively contributed to the A2A protocol and will support its implementation. They’ve also significantly expanded their Google Cloud practices with new experts and technical resources over the past year. This means our customers now have access to a global community of highly-trained AI experts who can help them develop AI agent use cases and strategies with interoperability in mind; prototype new applications; train and manage AI models; and ultimately deploy AI agents across their businesses.

Simplifying the ways we help partners go to market

Across the board, for all types of partners, Google Cloud is increasing resources to help them address customer demand for AI and cloud implementation services — and to better help partners identify the largest, most strategic opportunities to grow their businesses. This includes a 2X increase in funding for AI opportunities over the past year alone, and builds on our existing, nine-figure investment in partner learning that we’ve made over the past four years. 

In addition, we are continuing to refine the way we go-to-market with partners, to ensure partners have access to the right opportunities, and customers have access to the best possible technology and expertise. This includes:

• Better field alignment and co-sell: Google Cloud has been co-selling with partners since our inception. Now, we’re introducing new processes to better capture and share partners’ critical contributions with our sales team. This includes increased visibility into the highly valuable co-selling activities like workshops, assessments, and proofs-of-concept, as well as partner-delivered services for migrations, application modernization, and other managed services. Ultimately, this information will better enable our sales team to connect customers with the correct ISV and services partners.
• More partner earnings: We are continuing to evolve our incentives that help partners capitalize on the biggest opportunities, such as a 2x increase in partner funding for AI opportunities over the past year. We’re also introducing new AI-powered capabilities in Earnings Hub, our destination for tracking incentives and growth, which will enable partners to benchmark performance against peers, receive personalized tips on how to boost earnings, and more. 
• Supporting partner readiness: In the past four years, we’ve invested more than $100 million in partner training, and we are continuing to expand these efforts to help our ecosystem develop expertise in critical areas like Google Agentspace, Workspace, and more.

Making Google Workspace the best AI productivity platform for partners

More than 3 billion users and over 11 million paying customers rely on Google Workspace. Now, new enterprise-grade AI capabilities with Gemini are helping users get work done more quickly in tools like Gmail, Meet, Docs, and more. Our partners play a critical role in helping organizations around the world deploy Workspace, including providing critical migration support and training for organizations big and small around the world. 

This week at Next, we’re announcing new AI innovations in Workspace, including proactive analysis and visualization capabilities in Sheets, audio generation in Docs, a new way to automate work with AI agents in the loop, and more—all of which will make Workspace an even more attractive platform for customers and ensure partners are going to market with category-defining products. In addition, we’ve increased partner funding for Workspace opportunities by 4x over the past year, to ensure partners have the right resources and incentives to bring Workspace to customers.

With simplified pricing, an enhanced Gemini-powered feature set, and the ability to integrate Workspace with customers’ existing tools like Slack, we’re creating new and exciting opportunities for partners to create long-term, strategic engagements by implementing Workspace for customers. 

Announcing 2025 partner awards winners

In closing, we’re honored to spotlight this year’s winners of Google Cloud’s partner awards, which recognize the innovation and value that partners have created for customers—particularly with AI. Our ecosystem continues to evolve to meet the needs of businesses across industries, and we’re proud of the ways they have deployed Google Cloud’s technology to address complex challenges faced by our customers. To learn more about this year’s winners, please read the complete list. 

I’m excited to meet with thousands of partners this week and share ideas for how we can work together to support customers, and how we can provide a simple, effective go-to-market motion with our ecosystem. See you at Next ‘25!

^{1. IDC InfoBrief, sponsored by Google Cloud, Google Cloud AI: Driving Opportunity and Growth for Global Consulting & Systems Integrator Partners, doc #US53276025, April 2025}

We recently announced Gemini 2.5, our most intelligent AI model yet. Gemini 2.5 models are now thinking models, capable of reasoning before responding, resulting in dramatically improved performance. This transparent step-by-step reasoning is crucial for enterprise trust and compliance.

Our first model in this family, Gemini 2.5 Pro, available in public preview on Vertex AI, is now among the world’s best models for coding and tasks requiring advanced reasoning. It has state-of-the-art performance on a wide range of benchmarks, is recognized by many users as the most enterprise-ready reasoning model, and is at the top of the LM arena leaderboard by a significant margin. 

Building on this momentum, we are launching Gemini 2.5 Flash, our workhorse model with low latency and cost efficiency on Vertex AI, our comprehensive platform for building and managing AI applications and agents, and Google AI Studio.

Let’s dive into how these capabilities are transforming AI development on Google Cloud.

Advancing enterprise problem-solving with deep reasoning

Enterprises face challenges that require intricate information landscapes, multi-step analyses, and making nuanced decisions – tasks demanding that AI doesn’t just process, but also reasons. For these situations, we offer Gemini 2.5 Pro on Vertex AI, engineered for maximum quality and tackling the most complex tasks demanding deep reasoning and coding expertise. Coupled with a one million token context window, Gemini 2.5 Pro performs deep data analysis, extracts key insights from dense documents like legal contracts or medical records, and handles complex coding tasks by comprehending entire codebases.

“At Box, we’re redefining how enterprises apply intelligence to their content. With Box AI extract agents, powered by Gemini, users can instantly streamline tasks by making unstructured data actionable, as seen in millions of extractions supporting a variety of use cases, including procurement and reporting. Gemini 2.5 represents a leap forward in advanced reasoning, enabling us to envision building more powerful agent systems where extracted insights automatically trigger downstream actions and coordinate across multiple steps. This evolution pushes the boundaries of automation, allowing businesses to unlock and act upon their most valuable information with even greater impact and efficiency.”  — Yashodha Bhavnani, VP of AI Product Management, Box

“Moody’s leverages Gemini’s advanced reasoning capabilities on Vertex AI within a model-agnostic framework. Our current production system uses Gemini 2.0 Flash for intelligent filtering and Gemini 1.5 Pro for high-precision extraction, achieving over 95% accuracy and an 80% reduction in processing time for complex PDFs. Building on this success, we are now in the early stages of testing Gemini 2.5 Pro. Its potential for deeper, structured reasoning across extensive document sets, thanks to features like its large context window, looks very promising for tackling even more complex data challenges and enhancing our data coverage further. While it’s not in production, the initial results are very encouraging.” — Wade Moss, Sr. Director, AI Data Solutions, Moody’s

To tailor Gemini for specific needs, businesses can soon leverage Vertex AI features like supervised tuning (for unique data specialization) and context caching (for efficient long context processing), enhancing performance and reducing costs. Both these features are launching in the coming weeks for Gemini 2.5 models.

Building responsive and efficient AI applications at scale

While Gemini 2.5 Pro targets peak quality for complex challenges, many enterprise applications prioritize speed, low latency, and cost-efficiency. To meet this need, we will soon offer Gemini 2.5 Flash on Vertex AI. This workhorse model is optimized specifically for low latency and reduced cost, delivering impressive and well-balanced quality for high-volume scenarios like customer service or real-time information processing. It’s the ideal engine for responsive virtual assistants and real-time summarization tools where efficiency at scale is key.

Gemini 2.5 Flash will also feature dynamic and controllable reasoning. The model automatically adjusts processing time (‘thinking budget’) based on query complexity, enabling faster answers for simple requests. You also gain granular control over this budget, allowing explicit tuning of the speed, accuracy, and cost balance for your specific needs. This flexibility is key to optimizing Flash performance in high-volume, cost-sensitive applications.

“Gemini 2.5 Flash’s enhanced reasoning ability, including its insightful responses, holds immense potential for Palo Alto Networks, including detection of future AI-powered threats and more effective customer support across our AI portfolio. We are focused on evaluating the latest model’s impact on AI-assistant performance, including its summaries and responses, with the intention of migrating to this model to unlock its advanced capabilities.” — Rajesh Bhagwat, VP of Engineering, Palo Alto Networks

Optimizing your experience on Vertex AI

Choosing between powerful models like Gemini 2.5 Pro and 2.5 Flash depends on your specific needs. To make it easier, we’re introducing Vertex AI Model Optimizer in experimental, to automatically generate the highest quality response for each prompt based on your desired balance of quality and cost. For customers who have workloads that do not require processing in a specific location, our Vertex AI Global Endpoint provides capacity-aware routing for our Gemini models across multiple regions, maintaining application responsiveness even during peak traffic or regional service fluctuations.

Powering the future with sophisticated agents and multi-agent ecosystems

Gemini 2.5 Pro’s advanced multimodal reasoning enables sophisticated, real-world agent workflows. It interprets visual context (maps, flowcharts), integrates text understanding, performs grounded actions like web searches, and synthesizes diverse information – allowing agents to interact meaningfully with complex inputs.

Building on this potential, today we are also announcing a number of innovations in Vertex AI to enable multi-agent ecosystems. One key innovation supporting dynamic, real-time interactions is the Live API for Gemini models. This API allows agents to process streaming audio, video, and text with low latency, enabling human-like conversations, participation in live meetings, or monitoring real-time situations (such as understanding spoken instructions mid-task).

Key Live API features further enhance these interactions: support for long, resumable sessions (greater than 30 minutes), multilingual audio output, time-stamped transcripts for analysis, dynamic instruction updates within sessions, and powerful tool integrations (search, code execution, function calling). These advancements pave the way for leveraging models like Gemini 2.5 Pro in highly interactive applications.

Get started 

Ready to tackle complex problems, build efficient applications, and create sophisticated AI agents? Try Gemini 2.5 on Vertex AI now!

Today at Google Cloud Next, we’re thrilled to announce Firestore with MongoDB compatibility, built from the ground up by Google Cloud. It provides developers with an additional choice for their demanding document database workloads.

MongoDB compatibility has been a highly-requested capability from Firestore’s existing community of over 600,000 active developers. With this launch, Firestore developers can now take advantage of MongoDB’s API portability along with Firestore’s differentiated serverless service, to enjoy multi-region replication with strong consistency, virtually unlimited scalability, industry-leading high availability of up to 99.999% SLA, and single-digit milliseconds read performance. 

When complemented with the ability to use their existing MongoDB application code, drivers, tools, in addition to the open-source ecosystem of MongoDB integrations, Firestore developers are able to quickly build applications for common use cases, including content management systems, e-commerce product catalogs, and user profiles.

Firestore with MongoDB compatibility also offers a customer-friendly serverless pricing model, with no up-front commitments required. Customers only pay for what they use without the hidden costs of capacity planning. You can learn more about how to get started at our Firestore with MongoDB compatibility site.

“After migrating to Firestore, we improved developer productivity by 55%, observed better service reliability, and have been able to seamlessly scale to over 250,000 requests per second and 30 billion documents. Because Firestore is completely serverless and provides virtually unlimited scalability, we no longer have to worry about managing our underlying database infrastructure — liberating us from database DevOps. This has enabled us to focus on product innovations that matter to our customers,” said Karan Agarwal, director of engineering, HighLevel.

Here’s how Firestore with MongoDB compatibility is different

Developers enjoy the agility of the popular MongoDB API and query language to store and query semi-structured Javascript Object Notation (JSON) data. With this announcement, we’re implementing the MongoDB API natively in the existing Firestore service, allowing developers to use their MongoDB drivers and integrations to read and write to Firestore with no application code changes. Developers can now build their applications, enjoying the best of the MongoDB ecosystem, and leveraging Firestore’s experience in designing and managing the most scalable and available serverless document database service.

Benefits of Firestore with MongoDB compatibility:

1. Scale while maintaining performance, with zero intervention and downtime

At Firestore’s core is an intelligent, serverless document database service that serves some of the most demanding workloads in the world, powering more than 1.5 billion monthly active end-users. Firestore offers virtually unlimited horizontal scaling with zero customer intervention or downtime. 

We’re able to do this through a differentiated backend that offers automatic, real-time rebalancing of disaggregated compute and storage that smooths out load across nodes. It allows us to add resources exactly where they are needed. We’re now able to bring the best of Google Cloud to the document databases ecosystem.

2. Industry-leading availability

Firestore enables automatic, synchronous replication across different availability zones and regions. When any replica becomes unhealthy, Firestore will fail over to another replica with zero downtime and zero data loss. At the same time, it will apply automatic self-healing on the unhealthy replica. Unhealthy replicas will not affect processes such as automatic scaling.

3. Integrated Google Cloud governance

Firestore is fully integrated into Google Cloud’s billing, security, monitoring, auditing and compliance checks, which allows administrators to manage Firestore with other Google Cloud services through shared, central services such as Identity and Access Management (IAM), Cloud Monitoring, Cloud Logging, Database Center, and Assured Workloads.

4. Transparent, simple pricing

Firestore makes keeping costs in check easier than ever. Pricing is transparent, predictable and simple. For read and write operations conducted on the database, customers simply pay for the actual operations conducted, based on the size of documents and index entries in 4 kilobyte chunks for reads and 1 kilobyte chunks for writes. 

There are no upfront fees or hidden costs due to challenging cluster capacity planning, mismanaged cluster sharding, or I/O charges. Customers can also attain further discounts on the pricing on the operations conducted through one-year and three-year committed-use discounts. Storage is billed for actual storage consumed, and is inclusive of automatic data replication. Customers can explore examples of applied pricing here.

5. Maximize developer flexibility

Firestore offers developers more interface choices. Coming soon, we will also offer data interoperability between Firestore’s MongoDB compatible interface and Firestore’s innovative real-time and offline SDKs. This can allow developers to maximize existing libraries and tools from both of the MongoDB and Firestore developer communities.

Get started on Firestore with MongoDB compatibility

With this launch, Google Cloud is offering developers seeking a MongoDB interface multiple choices, including both MongoDB Atlas and Firestore. We’re thrilled to see what you’ll be able to achieve using Firestore with MongoDB compatibility. Firestore with MongoDB compatibility is available in preview as part of the new Firestore Enterprise edition. Get started today on Firestore with no upfront fees and a free tier.