Skip to main content
Cloud / Google Cloud / Products / Deployments (formerly Agent Engine) - AI Agent Runtime

Deployments (formerly Agent Engine) - AI Agent Runtime

Deployments/Agent Runtime (formerly Vertex AI Agent Engine) is Google Cloud's managed runtime for autonomous AI agents in production.

AI/ML
Pricing Model Pay-per-use (compute time + invocations)
Availability Global, EU regions
Data Sovereignty EU regions available
Reliability SLA as published by the provider SLA

Deployments, also referred to as Agent Runtime, is Google’s managed runtime environment for operating autonomous AI agents in production. The service evolved from Vertex AI Agent Engine, introduced in 2024, and is now part of the Gemini Enterprise Agent Platform (formerly Vertex AI). It targets development teams that want to run agent frameworks such as the Agent Development Kit (ADK), LangChain, LangGraph, or LlamaIndex at scale without managing their own infrastructure.

What is Deployments (Agent Runtime)?

Deployments handles all the infrastructure surrounding the operation of LLM agents: automatic scaling, session and memory management via a Memory Bank, logging/tracing, and integration into the Google Cloud ecosystem. Developers deploy their existing agent frameworks directly onto the runtime without needing to manage containers or Kubernetes clusters themselves. The service supports tool use — the ability of agents to call external functions and APIs — and manages the associated state across multiple conversation turns.

An important distinction from the related Agent Studio (formerly Vertex AI Agent Builder): Agent Studio focuses on low-code creation of RAG applications, search systems, and chatbots via a graphical interface. Deployments/Agent Runtime, on the other hand, is the programmatic runtime environment for code-first agents that execute complex workflows, orchestrate multiple tools, and integrate into existing systems.

The service offers resource controls (CPU, memory, concurrency limits), custom service accounts and agent identities, and monitoring via traces and logs. Integration with other Google Cloud services lets agents access enterprise data directly. Security features such as VPC Service Controls and IAM-based access control make the service production-ready for enterprise applications.

Integration with innFactory

As a certified Google Cloud partner, innFactory supports you in designing and operating AI agent architectures on Deployments/Agent Runtime — from choosing the right agent framework and integrating tools to production-ready deployment on the Gemini Enterprise Agent Platform.

Contact us for a consultation on Deployments/Agent Runtime and autonomous AI systems.

Typical Use Cases

Production operation of LLM agents
Multi-agent systems
Tool use and function calling for AI
Orchestration of agent workflows

Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of Google Cloud (official documentation). This page does not represent an offer by Google Cloud.

Google Cloud Partner

innFactory is a certified Google Cloud Partner. We provide expert consulting, implementation, and managed services.

Google Cloud Partner

Similar Products from Other Clouds

Other cloud providers offer comparable services in this category. As a multi-cloud partner, we help you choose the right solution.

STACKIT

STACKIT AI Model Experiments: Managed MLflow

STACKIT AI Model Experiments: managed MLflow for experiment tracking, LLM tracing, and EU AI Act audit trails on EU …

Pricing Public Preview: service itself free, …
Compare →
STACKIT

STACKIT AI Model Serving: Sovereign LLMs

STACKIT AI Model Serving: Run open-weight LLMs like Llama, Qwen, and GPT-OSS GDPR-compliant from German data centers, …

Pricing Pay-as-you-go per token (input/output)
SLA Runs on the data-sovereign STACKIT Cloud
Compare →
STACKIT

STACKIT Dremio - Sovereign Data Lakehouse

STACKIT Dremio: managed data lakehouse based on Dremio for SQL queries without data movement. GDPR-compliant, public …

Pricing Public preview; unified billing via …
SLA SLA as published by the provider; the service is currently in public preview
Compare →
STACKIT

STACKIT Intake - Data Ingestion for the Data Lakehouse

STACKIT Intake: managed data ingestion via the Kafka protocol directly into Apache Iceberg tables for the STACKIT Data …

Pricing Capacity-based; capacity is configured …
SLA SLA as published by the provider
Compare →
STACKIT

STACKIT Notebooks - Managed JupyterHub

STACKIT Notebooks: managed JupyterHub/JupyterLab for data science and ML in German and Austrian data centers.

Pricing Pay-per-use (compute resources of the …
SLA SLA as published by the provider
Compare →
STACKIT

STACKIT Workflows - Managed Apache Airflow

STACKIT Workflows is a managed workflow orchestration service built on Apache Airflow for data pipelines and ML …

Pricing Pay-per-use
SLA SLA as published by the provider
Compare →

99 comparable products found across other clouds.

Ready to start with Deployments (formerly Agent Engine) - AI Agent Runtime?

Our certified Google Cloud experts help you with architecture, integration, and optimization.

Schedule Consultation