OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Infrastructure & Serving

Genai Factory

Genai Factory is a public collection of Terraform blueprints for deploying generative AI infrastructure and sample applications on Google Cloud. Its factories separate cloud project and networking prerequisites from application resources, and include patterns such as Cloud Run agents, retrieval-augmented generation and natural-language querying of BigQuery.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Genai Factory — OpenViST project profile in AI Infrastructure & Serving.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when your team needs a Terraform starting point for deploying a supported generative AI pattern on Google Cloud and can provide the required project, billing, permissions and networking setup.

Who it is for

  • Google Cloud platform and infrastructure teams deploying generative AI environments.
  • Developers who want to extend sample infrastructure deployments with their own application code.
  • Organizations using Cloud Foundation Fabric or existing Google Cloud projects and networking.

Documented uses

  • Provision Google Cloud projects, service accounts, IAM permissions and networking for an AI deployment.
  • Deploy a selected generative AI platform and sample application through Terraform.
  • Start from a documented pattern for Cloud Run, RAG, agent runtimes or natural-language BigQuery queries.
Before you choose

Requirements & limits

The repository is public and licensed under Apache-2.0.

  • Terraform is the deployment mechanism, with Cloud Foundation Fabric modules and Google Cloud Terraform resources.
  • The optional 0-prereqs stage creates foundational projects, identities, permissions and networking; 1-apps deploys platform resources and the sample application.
  • The quickstart assumes project-management permissions and permission to link a billing account.
  • Cloud Foundation Fabric compatibility is specified from v59.0.0; compatibility with master is not guaranteed.
  • Existing projects and networking can be used, but the 1-apps stage requires their relevant configuration details.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare a Cloud Run factory deployment

Use the documented quickstart stages to provision prerequisites and deploy platform resources for the Cloud Run Single factory.

Before you start

  • Permissions to create and manage Google Cloud projects and link them to a billing account.
  • A selected factory directory; the documented example is cloud-run-single.
  • Terraform.

Steps

  1. Enter the cloud-run-single directory, then open 0-prereqs.

  2. Copy terraform.tfvars.sample to terraform.tfvars and replace the prefix, billing account and parent values.

  3. Run terraform init and terraform apply in 0-prereqs.

  4. Open 1-apps, copy its terraform.tfvars.sample to terraform.tfvars and customize it; then run terraform init and terraform apply.

  5. Deploy the application by following the commands printed in the output.

Expected outcome

The prerequisites and platform resources for the selected factory are applied, followed by the application deployment stage.

Guide limitations

  • The quickstart's prerequisite stage is optional if you can create projects and manage shared VPCs.
  • If projects are created outside the prerequisite stage, the 1-apps stage needs the required projects, APIs, service accounts, roles and networking information.
  • The guide documents stages and commands, not a verified successful deployment.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
GoogleCloudPlatform/genai-factory · Official homepage title · retained in its source language
How we review sourcesSuggest a correction