Google Cloud Announces Gemini Agent
- •Google Cloud announced Gemini Agent, a multi-agent tool that runs in the cloud.
- •The service supports Gemini 4 Argon, Claude Opus 5.5, and Claude Sonnet 5.5.
- •More models are planned, while pricing and availability in Japan remain undisclosed.
Google Cloud announced Gemini Agent on October 9, 2026 (local time), describing it as a cloud-based tool that uses multiple AI agents.
According to an explanation posted on X by Shubham Sabu of Google, Gemini Agent is not tied to a single model and uses the model best suited to each task. The models available at launch are Google’s Gemini 4 Argon and Anthropic’s Claude Opus 5.5 and Claude Sonnet 5.5. Sabu said other closed and open models are planned for addition.
At the time of the announcement, Google had not explained how Gemini Agent relates to the existing Gemini app or Gemini Enterprise. It had also not said which individual Gemini AI, Google Workspace, or Gemini Enterprise plans would include the service.
ITmedia AI+ said it had asked Google Cloud Japan about the service’s relationship to existing products, pricing, and plans for availability in Japan, and would publish more details after receiving a response. Those details, along with the service’s access conditions and launch timing, were not announced at publication. The planned additional models were not named.