Gemini Enterprise Agent Platform

Build, scale, govern, and optimize with Gemini Enterprise Agent Platform

Gemini Enterprise Agent Platform is our open and comprehensive platform that empowers businesses to rapidly build, scale, govern, and optimize enterprise-grade agents grounded in your enterprise data. It provides the full-stack foundation and extensive developer choice you need to transform your applications and workflows into powerful, reliable agentic systems at global scale.

API quickstart Get started in Agent Studio Get an API key

Everything you need to buildproduction-ready agents

Build

  • Build with Agent Platform
    Learn about the different tools and services you can use to build agents with Agent Platform.
  • Agent Development Kit
    The Agent Development Kit (ADK) is a modular, model-agnostic framework designed to make building and deploying complex AI agents feel just like standard software development.
  • Agent Garden
    Explore Google's library of pre-built agents.
  • Agent Studio
    Low-code canvas for designing complex multi-agent reasoning loops.
  • Skill Registry
    Skill Registry serves as a secure, private, and low-latency repository for managing and discovering agent skills.
  • RAG Engine
    RAG Engine helps you securely connect your private data to LLMs, reducing hallucinations and improving answer accuracy by providing the model with relevant, context-rich information.
  • Vector Search
    Vector Search is an AI-native search engine that combines storage and retrieval into one scalable, auto-tuned platform, simplifying how you store, search, and manage your data for AI applications.
  • Managed Agents API
    A config-driven, REST-first API for building autonomous agents inside a fully managed sandbox.

Scale

  • Deploy agents
    Deploy agents on Agent Engine using Python SDK, source files, or container images.
  • Manage deployed agents
    Manage and update deployed agents, including scaling and resource allocation.
  • Use an agent
    Interact with deployed agents using the API or SDK.
  • Agent Platform Sessions
    Manage conversational sessions and state for multi-turn interactions.
  • Agent Platform Memory Bank
    Store and retrieve long-term memories for agents to maintain context.
  • Set up Memory Bank
    Configure and initialize the Memory Bank for your agents.
  • Generate memories
    Extract and save memories from agent interactions.
  • Fetch memories
    Retrieve stored memories to provide context to agent executions.

Govern

  • Agent Gateway
    Centralized gateway for routing, securing, and monitoring agent traffic.
  • Semantic governance
    Configure policies to ensure safe and compliant agent behavior.
  • Content security
    Monitor and filter agent inputs and outputs for safety and security.
  • Identity & IAM
    Manage access control and permissions for agents using Google Cloud IAM.
  • Share agent
    Securely share agents across projects or organizations.

Optimize

  • Agent evaluation
    Learn about evaluation concepts and how to assess agent performance.
  • Observability overview
    Monitor agent health, latency, and resource usage.
  • Traces
    Debug complex reasoning loops and tool calls with detailed execution traces.
  • Topology
    Visualize agent architectures and dependency graphs.
  • Offline evaluations
    Run evaluations on static datasets to measure quality and latency.
  • Simulated evaluation
    Simulate user interactions and evaluate agent behavior against custom criteria.
  • Online monitoring
    Continuously monitor and evaluate agent performance in production environments.
  • Optimize prompts
    Auto-tune and optimize prompts to improve agent response quality and efficiency.

Built for developers,designed for enterprise

  • Gemini
    Advanced reasoning-first models optimized for complex agentic workflows and coding. Features Gemini 3.1 Pro and Gemini 3 Flash.
  • Veo
    High-quality video generation models that create cinematic video from text or image prompts. Features Veo 3.1.
  • Lyria
    Generate full-length music tracks and audio clips from simple text and image prompts. Features Lyria 3.
  • Partner models
    Access curated third-party models offered as managed APIs. Includes Claude, Grok, and Mistral.
  • Open models
    Choose from a variety of open weight models and select the best serving option. Includes Llama, DeepSeek, Mistral, and Qwen.

Learn more about Agent Platform

For high-level product overviews and marketing details across Google Cloud products, visit the Gemini Enterprise Agent Platform product page.

Go to the Agent Platform product page