Gemini Enterprise Agent Platform

What it does

Gemini Enterprise Agent Platform is a Google Cloud environment for building, deploying, governing, and monitoring AI agents that use reasoning and tools to automate workflows, search across enterprise systems, and generate content. Teams can work with low-code Agent Studio, the code-first Agent Development Kit, or a managed Agents API, and can choose from over 200 foundation models in Model Garden, including Google's Gemini models and third-party and open models.

In practice, developers deploy agents to a managed Agent Runtime with session history and optional long-term Memory Bank. Agents receive a unique Agent Identity, traffic is governed by Agent Gateway, and tool access is centralized in an Agent Registry. Cloud Observability collects traces, logs, and metrics by default when agents are deployed with ADK, which helps debug and track token usage.

The platform suits technical teams and enterprises that already run on Google Cloud and need governance, IAM controls, and auditing. Access requires a Google Cloud project and billing account, and costs are usage-based across tokens, runtime compute, memory, and storage. Pricing includes a $300 new-customer credit, and the documentation notes that legacy AI Platform features have differences such as no scale-to-zero for inference.

What it can do

• Build agents with low-code Agent Studio or code-first Agent Development Kit
• Access over 200 Google, partner, and open-source models through Model Garden
• Deploy agents to a fully managed Agent Runtime with session management
• Provide long-term personalized memory across sessions with Agent Platform Memory Bank
• Govern agents with Agent Identity, Agent Registry, and Agent Gateway policy enforcement
• Monitor agent traces, logs, and metrics with Cloud Observability and evaluate quality with Gen AI evaluation

Works with

• Google Cloud IAM
• BigQuery
• Google Workspace Admin Console
• Gemini Enterprise

Example workflows

Customer support agent

Input: support tickets and a connected knowledge base. Steps: build the agent in Agent Studio, attach tools through the Registry, deploy to Agent Runtime, and review traces in Cloud Observability. Output: an agent that drafts responses to common inquiries, subject to the tools and data it is granted.

Cross-system information discovery

Input: natural-language questions about project status across internal systems. Steps: build the agent with ADK, grant it Agent Identity permissions, and use Agent Gateway to control tool calls. Output: answers assembled from permitted sources, with an audit trail of agent activity.

Before you choose

• Costs are usage-based across model tokens, agent compute, memory, storage, and optional services such as Model Armor, so estimates require the pricing calculator or sales contact.
• Documentation requires IAM and Agent Gateway configuration for governance; teams without Google Cloud experience will need a learning curve.
• Legacy AI Platform inference features such as scale-to-zero and some lower-cost machine types are not available in the current platform.

Listed under

• Coding
• AI agents
• Developers
• Enterprise
• Gemini API
• Machine learning
• MLOps

Get started

Start free trial with new-customer credit, then build in Agent Platform.

1. Create or select a Google Cloud project and enable billing; new customers can start with a free trial.
2. Choose a development path: Agent Studio for low-code, Agents API for config-driven agents, or ADK for custom code.
3. Deploy the agent to Agent Runtime and assign it an Agent Identity with the least permissions it needs.

Model usage is billed per token through pay-as-you-go tiers or Provisioned Throughput subscriptions. Agent infrastructure is metered by vCPU, RAM, and storage. New customers receive a $300 credit valid for 90 days, plus monthly Free Tier allowances of 50 vCPU-hours, 100 GiB-hours of RAM, and 1 GiB storage. Rates for Agent Studio and the Agents API are not confirmed in the reviewed information.

90-day free trial.

Source: cloud.google.com/products/gemini-enterprise-agent-platform, checked 28 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Gemini platform | Google Cloud ↗cloud.google.com
• Gemini platform pricing | Google Cloud ↗cloud.google.com
• Agents overview  |  Gemini Enterprise Agent Platform  |  Google Cloud Documentation ↗docs.cloud.google.com

Company website

cloud.google.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.