The company made the announcement on Thursday, describing the new system as a big step forward from Gemini 3, which it introduced in November 2025. It is already rolling out the preview across consumer, developer and enterprise products.

In a statement, Google said: “Last week, we released a major update to Gemini 3 Deep Think to solve modern challenges across science, research and engineering. 

“Today, we’re releasing the upgraded core intelligence that makes those breakthroughs possible: Gemini 3.1 Pro. We are shipping 3.1 Pro across our consumer and developer products to bring this progress in intelligence to your everyday applications.”

Developers can now access Gemini 3.1 Pro in preview through the Gemini API in Google AI Studio, Gemini CLI, Google Antigravity and Android Studio.

Enterprise customers can use it in Vertex AI and Gemini Enterprise. Consumers are getting it through the Gemini app and NotebookLM.

Google says the new model improves core reasoning. It reports that Gemini 3.1 Pro scored 77.1 per cent on ARC-AGI-2, a benchmark designed to test how well a system can solve entirely new logic patterns. According to the company, that score is more than double the reasoning performance of Gemini 3 Pro.

On another benchmark known as Humanity’s Last Exam, Google says Gemini 3.1 Pro outperformed Anthropic’s Claude Opus 4.6 and OpenAI’s GPT-5.2. Independent testing platform APEX-Agents also ranked the model first for professional task performance.

Brendan Foody, chief executive of Mercor, which runs the APEX system, wrote on social media: “Gemini 3.1 Pro is now at the top of the APEX-Agents leaderboard,” adding that the model’s results show “how quickly agents are improving at real knowledge work.”

Google describes Gemini 3.1 Pro as a stronger base for complex problem-solving. It says the system can support tasks that require detailed reasoning, data synthesis and multi-step workflows.

The company appears to be focusing on tools that go beyond simple text generation and can handle layered instructions.

The update comes just three months after the release of Gemini 3. In that same period, competitors including OpenAI and Anthropic have also introduced new models, intensifying competition among large technology firms.

Google DeepMind has published a model card for Gemini 3.1 Pro outlining its intended uses, limits and safety measures.

The company says the model is natively multimodal, meaning it can process text, images and other forms of input. While that expands its use cases, it also leads to safety and content moderation questions.

For now, Google says the preview will allow it to test improvements further before a general release. Higher usage limits are being introduced for users on the Google AI Pro and Ultra plans, while NotebookLM access is limited to those subscribers.

The company ended its announcement with a brief note to users: “We can’t wait to see what you build and discover with it.”

The collaboration is set to redefine how iconic brands such as Dove, Vaseline, and Hellmann’s engage with consumers in an increasingly digital and “agentic” commerce landscape.

The move signals a major shift for Unilever as it migrates its data and cloud infrastructure to Google Cloud to build an AI-first digital backbone.

Building an ‘AI-First’ Digital Backbone

Under the agreement, Unilever will leverage Google’s enterprise AI platform, Vertex AI, to develop new capabilities in brand discovery and marketing.

The goal is to move beyond traditional advertising toward conversational and agentic experiences, where intelligent systems interact with consumers and execute complex business tasks.

By transitioning its integrated data platform to Google Cloud, Unilever aims to:

• Accelerate Demand Generation: Respond to market shifts with higher agility by turning raw data into actionable insights.
• Deploy Agentic Workflows: Implement intelligent systems capable of managing complex processes across the value chain.
• Modernize Marketing: Utilize AI-augmented tools for better measurement and consumer conversion.

Redefining Value through Technology

The partnership highlights the growing importance of “sovereign” data and specialized AI models in the Fast-Moving Consumer Goods (FMCG) sector.

Willem Uijen, Unilever’s chief supply chain and operations officer, noted that technology is no longer a support function but the core of value creation.

“As brands are increasingly discovered and chosen in environments shaped by AI, we must lead this shift. This collaboration… ensures Unilever is agile, fit for the future, and equipped to unlock value at every level,” Uijen stated.

Tara Brady, president of Google Cloud EMEA, added that the deployment of models like Gemini will create a “system of intelligence” that reasons and learns, setting a new standard for consumer engagement.

The Three Pillars of Collaboration

The five-year roadmap is built on three strategic areas designed to maintain Unilever’s competitive edge:

Agentic Commerce: Building next-gen marketing capabilities to stay ahead of shifting consumer habits.

Integrated Cloud Foundation: Transitioning key enterprise applications to Google Cloud for scalable AI deployment.

Advanced AI Adoption: Combining Unilever’s CPG expertise with Google’s pioneering tech to sustain long-term growth.

This partnership is a clear indicator that the CPG industry is moving away from traditional push marketing toward AI-driven pull commerce.

For Unilever, the move to Google Cloud isn’t just a technical upgrade; it’s a defensive and offensive play to ensure its brands remain visible as AI agents, rather than humans, start making more purchasing decisions in the future.

But what stood out was the breadth, the partnerships, and the quiet confidence that Google Cloud is no longer trying to catch up, it has caught up and outpaced.

What’s the Big Idea?

In simple terms, Google Cloud wants to be the backbone of the AI-powered enterprise. Not just with flashy models, but with tools you can actually build on, systems that plug into what you’ve already got, and infrastructure that can handle the real-world mess of data, scale, and regulation.

At the heart of this innovation is Vertex AI, Google’s platform that now supports everything from multi-agent ecosystems to generative video, image, speech, and music creation. 

Gemini 2.5, the latest model, can reason before responding—a big one from pattern-matching to thinking. That’s muscle.

Who’s Jumping Onboard?

Some of the world’s biggest names are already building with Google Cloud. Freshfields, a top legal firm, is deploying Gemini to speed up due diligence. Reddit has handed part of its homepage experience to Vertex AI. Even Samsung’s cute little home robot, Ballie, now runs Gemini.

Mattel’s using it to spot trends. Papa Johns wants it to improve customer experience. And Lloyds Banking Group? They’re migrating major systems—yes, banking systems—to Google Cloud to unlock AI innovation. That’s a big deal.

Healthcare is Changing Too

It’s not all business-as-usual. Manipal Hospitals in India and Seattle Children’s Hospital are using Google Cloud tools to tackle healthcare challenges. They’re building AI agents that help clinicians access best practices quickly or design cancer models that learn from vast datasets. If this works as intended, it’s not just innovation—it’s impact.

The Quiet Power Play: Ironwood TPUs

Google also dropped Ironwood, its latest Tensor Processing Unit (TPU), built specifically for inference—what happens when models are used in real time. It’s paired with upgrades to Google’s AI Hypercomputer, enabling high-speed training and model deployment with tighter efficiency. It’s more than silicon; it’s the plumbing that makes the AI wave flow.

Let’s Talk About Money

Beyond the tech, Google Cloud is working to be the platform where startups, banks, hospitals, and governments alike can build. Not alone, but together—with global partners like Deloitte, Capgemini, and KPMG all using Google’s Agentspace to create AI agents tailored for industries.

There’s also a quiet arms race happening in interoperability. Google’s new Agent2Agent protocol aims to make AI agents talk to each other—across vendors, ecosystems, even clouds. That’s a nod to the future, a world of mixed AI systems, not one winner.

So, What’s the Takeaway?

This is about Google finally being ready to lead from the front—not just with promises, but with partnerships, usable tools, and infrastructure that actually works. The Google Cloud Next ’25 event made us know that AI isn’t coming. It’s here. And Google wants to build it with you, not for you.

If the execution matches the vision, we’re looking at a very different enterprise space by the time Next ’26 rolls around.