What Google Cloud announced in AI this month

Google Cloud introduced Gemini 4 Argon, a model designed for complex, multi-step enterprise tasks such as software builds, financial analysis, and cybersecurity workflows. Argon maintains context across long sequences and integrates domain-specific expertise to improve precision in specialized workflows. The model is positioned to support teams handling detailed, iterative processes that require sustained focus and accuracy.

Anthropic’s Claude models, including Opus 5.5 and Sonnet 5.5, are now available on Google Cloud through both Anthropic-managed and Google-managed deployments. Google Cloud also added OpenAI models, making it the only hyperscale cloud provider offering access to all three leading closed-source LLMs alongside over 200 first- and third-party models on a single platform. This expansion aims to give organizations greater flexibility in model selection and deployment.

Gemini for government services entered preview, offering purpose-built AI tools tailored to public sector workflows. The solution includes specialized skills for government tasks, integration with existing agency systems, and governance controls. Additionally, Google DeepMind’s Gemini Robotics Embodied Reasoning 2 (ER 2) is now available, enabling organizations to deploy safe, adaptive physical AI for robotics with enterprise-grade security and lifecycle management.

Developer resources included guides on implementing MCP apps in Gemini, evaluating agent plugins, and auditing LLM evaluators. New tools like the Google Cloud CLI remote MCP server and Graph Workflows in ADK were introduced to streamline infrastructure management and AI orchestration. Google also highlighted cost-control measures for specialized industry models in financial services and legal sectors.