shopify 5g2sFtzmax 1000x1000 1
At Google Cloud, we’re committed to providing customers with the leading selection of models to build and scale production-ready AI apps and agents on a platform optimized for performance, trust, and global scale. Today, we’re further expanding Vertex AI’s curated collection of models with the addition of Anthropic’s newest release: Claude Opus 4.6.
Claude Opus 4.6 is Anthropic’s most powerful model yet. In addition to excelling at complex coding tasks and creating sophisticated agents, Opus 4.6 delivers a step-change in enterprise workflows—generating documents, spreadsheets, and presentations with professional polish and domain awareness. Top use cases include:
| Feature | Claude Opus 4.6 on Vertex AI Availability |
|---|---|
| Adaptive Thinking | GA |
| Fine-grained tool streaming toggle | GA |
| Effort parameter | GA |
| 128k Output Tokens | GA |
| Compaction API | Preview |
| 1M Context Window | Preview |
| Tool params quoting consistency | GA |
Vertex AI eliminates the trade-off between accessing frontier models and utilizing a world-class AI development platform. Because building sophisticated agents requires more than just a model, Vertex AI provides a complete agentic stack designed to manage the complexities of production, governance, and global scale.
Rapidly build intelligent agents: Leverage the Vertex AI Agent Builder stack—including the Agent Development Kit—to move multi-step agentic workflows from prototype to scaled production.
Scale agents with fully managed services: Use Agent Engine to deploy and scale agents in a serverless environment. With Memory Bank and Sessions, your Claude-powered agents maintain persistent, long-term context across multiple interactions for a more natural user experience.
Improve performance and reduce costs: Maintain consistent performance during peak demand by reserving dedicated capacity at a fixed cost with Provisioned Throughput. To further improve efficiency, leverage features like the global endpoint for Claude, prompt caching with flexible Time To Live (TTL), batch predictions, web search (preview), and a 1M token context window (preview) for Claude Opus 4.6.
Run on optimized infrastructure: Vertex AI’s infrastructure is purpose-built to handle the compute demands of frontier models to help lower latency and improve efficiency.
Govern and secure at scale: Integrated governance controls keep agents on task and their actions logged. Secure the full AI stack including models and agents with Google Cloud’s foundational security and data residency controls, protect against novel AI threats like prompt injection and tool poisoning with Model Armor, and get advanced AI Protection in Security Command Center to prioritize AI risks and detect threats.
“The combination of Claude and Vertex AI helps us empower millions of merchants with our AI-enabled commerce assistant, Sidekick. There’s been a lot of merchants that we’ve talked to that say Sidekick ‘feels like a lifeline for small businesses’ and helps them ‘come out on top’.”
– Andrew McNamara, Director of Applied AI, Shopify
Read the case study
“Claude in combination with Vertex AI, delivers stability and reliability for our customers when it comes to building code for complex multi-step tasks and executing tool calls.”
– Amitay Gilboa, CEO, spring.new
Read the case study
“Getting a model as powerful as Anthropic’s Claude on Vertex AI is a win-win that makes life so much easier. We get a model that excels at tool calling on a comprehensive platform that integrates with our core Google Cloud workloads like GKE and Cloud Run — that’s the magic.”
– Justin Watts, Distinguished Engineer, TELUS
Read the case study
“Thanks to the quality of Anthropic’s technology and the ease of implementing it through Vertex AI, we switched our chat feature to Claude Sonnet 3.5 and improved performance in a single weekend.”
– Guy Gur-Ari, co-founder of Augment Code
Read the case study
“With Claude running on Vertex AI, we saw a 20% to 30% increase in code development velocity. Running Claude on Vertex AI not only accelerates development projects, it enables us to hardwire security into code before it ships.”
– Gunjan Patel, Director of Engineering, Office of the CPO at Palo Alto Networks
Read the case study
“Our AI agent is made more powerful through Claude models running on Vertex AI. This integration allows us to easily connect with other Google Cloud services, like Cloud Run, to work together behind the scenes to help customers turn their ideas into apps.”
– Amjad Masad, Founder and CEO, Replit
Read the case study
Claude Opus 4.6 is now generally available on Vertex AI. Enterprises can also quickly deploy the model through Google Cloud Marketplace, streamlining procurement and billing.
To begin your development journey, explore our sample notebook and visit our documentation for comprehensive pricing and regional availability details.
Just a few samples from a lora trained using Z image base. First 4 pictures…
AI agents that use tools, make decisions, and complete multi-step tasks aren't prototypes anymore.
This is a guest post co-written with David Meredith and Josh Zacharias from Associa. Associa,…
A team of geologists found for the first time evidence linking regions of low seismic…
In a typical online meeting, humans don't always wait politely for their turn to speak.…
Many people reported that the lora training sucks for z-image base. Less than 12 hours…