https://pgabriel-01.github.io/mlops-v3/scenarios/index.htmlPick one with scenario: in the inputs file. List them with Get-MLOpsScenario. Scaffold an inputs file with New-MLOpsInputFile -Scenario <id or number> -Path inputs.yaml -Iac <terraform or bicep> -Cicd <github or azdo>. # Id Network Serving Adds 1 basic-public public none Workspace, storage, Key Vault, Azure Container Registry, Application Insights, CPU cluster, training pipeline 2 managed-vnet-internet managed VNet, internet outbound online Private endpoints, public access disabled 3 managed-vnet-approved managed VNet, approved outbound only online FQDN and service tag allow-list (allowed_outbound_fqdns) 4 byo-vnet your VNet (created) online Spoke VNet, subnets, private DNS zones, private endpoints (vnet_address_space) 5 batch-scoring public batch Batch endpoint and deployment, nightly scoring workflow 6 online-bluegreen public online (blue/green) Deploy workflow shifts 10% then 100% of traffic and retires the old deployment 7 feature-store public online Managed feature store workspace 8 registry-promotion public online Shared Azure ML registry in the first environment, ml-promote workflow, deploy from the registry 9 governance public online Diagnostics, budget alerts (budget_amount, alert_emails), resource lock, policy 10 genai-foundry public none AI Foundry hub, AI Services account and project 11 existing-platform existing subnet and DNS online Reuses existing_platform subnet, private DNS zones and Log Analytics; governance on 12 rag-llmops public none AI Foundry plus Azure AI Search, model deployments, sample index and an evaluation quality gate Choosing a scenario Start here if… Scenario Relative fixed cost You are learning or prototyping and data is not sensitive 1 basic-public Lowest (about $60 per month for 3 environments) You need real-time scoring with a REST endpoint 6 online-bluegreen (safe rollouts) or 2 to 4 (private) Low to medium Scoring runs on a schedule, not in real time 5 batch-scoring Low (compute only while running) Security requires no public access, and you have no existing network 2 managed-vnet-internet, or 3 for locked-down egress Medium; 3 adds a managed firewall, which is the largest fixed cost You must control the network yourself 4 byo-vnet Medium (private endpoints and DNS zones) A platform team already owns the subnet, DNS and Log Analytics 11 existing-platform Medium; reuses shared services Several teams share models across workspaces or environments 8 registry-promotion Low added cost You need reusable features across models 7 feature-store Low added cost Finance or audit needs budgets, policy and diagnostics 9 governance Low added cost You are building generative AI apps 10 genai-foundry Depends on the models you deploy You are building a chat app over your own documents with an evaluation gate 12 rag-llmops Adds Azure AI Search (basic tier is roughly 75 USD per month) plus model token use Costs are relative and exclude training compute and endpoint instance hours, which dominate once you deploy. The default is three environments (see the parameter checklist). Set two to cut the fixed cost.Hugoen-ushttps://pgabriel-01.github.io/mlops-v3/scenarios/01-basic-public/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/01-basic-public/index.htmlWorkspace, storage, Key Vault, Azure Container Registry, Application Insights and a compute cluster. Training pipeline only, with public endpoints. Id basic-public Network public Serving none Adds Workspace, storage, Key Vault, Azure Container Registry, Application Insights, CPU cluster, training pipeline IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Use it New-MLOpsInputFile -Scenario basic-public -Path inputs.yaml -Iac terraform -Cicd github See the cost estimates and the prerequisites.https://pgabriel-01.github.io/mlops-v3/scenarios/02-managed-vnet-internet/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/02-managed-vnet-internet/index.htmlWorkspace managed virtual network (AllowInternetOutbound), private endpoints and public access disabled. Id managed-vnet-internet Network managed VNet, internet outbound Serving online Adds Private endpoints, public access disabled IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Public network access is disabled on the workspace and its dependencies. Outbound traffic from jobs goes to the internet through the managed network. Use it New-MLOpsInputFile -Scenario managed-vnet-internet -Path inputs.yaml -Iac terraform -Cicd github Note Pipelines that touch the workspace need self-hosted agents or runners with network access to it.https://pgabriel-01.github.io/mlops-v3/scenarios/03-managed-vnet-approved/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/03-managed-vnet-approved/index.htmlWorkspace managed virtual network (AllowOnlyApprovedOutbound) with an FQDN and service tag allow-list. Id managed-vnet-approved Network managed VNet, approved outbound only Serving online Adds FQDN and service tag allow-list (allowed_outbound_fqdns) IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Outbound traffic is limited to the allow-list. A managed Azure Firewall is created per environment, which is the largest fixed cost of any scenario. Use it New-MLOpsInputFile -Scenario managed-vnet-approved -Path inputs.yaml -Iac terraform -Cicd github Inputs to review allowed_outbound_fqdns: see Options Note Pipelines that touch the workspace need self-hosted agents or runners with network access to it.https://pgabriel-01.github.io/mlops-v3/scenarios/04-byo-vnet/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/04-byo-vnet/index.htmlSpoke VNet, subnets, private DNS zones and private endpoints for all workspace dependencies. Id byo-vnet Network your VNet (created) Serving online Adds Spoke VNet, subnets, private DNS zones, private endpoints (vnet_address_space) IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps You own the network. The accelerator creates a spoke VNet, subnets, private DNS zones and private endpoints. It creates no network security groups on the subnets, so add them to match your standards.https://pgabriel-01.github.io/mlops-v3/scenarios/05-batch-scoring/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/05-batch-scoring/index.htmlCompute cluster plus a batch endpoint and deployment with a scheduled scoring job. Id batch-scoring Network public Serving batch Adds Batch endpoint and deployment, nightly scoring workflow IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Scoring runs as a scheduled batch job (a nightly workflow) instead of a real-time endpoint. Use it New-MLOpsInputFile -Scenario batch-scoring -Path inputs.yaml -Iac terraform -Cicd github See the cost estimates and the prerequisites.https://pgabriel-01.github.io/mlops-v3/scenarios/06-online-bluegreen/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/06-online-bluegreen/index.htmlManaged online endpoint. The deploy workflow ships a new deployment and shifts traffic gradually. Id online-bluegreen Network public Serving online (blue/green) Adds Deploy workflow shifts 10% then 100% of traffic and retires the old deployment IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps The deploy workflow creates a new deployment, shifts 10% then 100% of traffic, and retires the old one. Use it New-MLOpsInputFile -Scenario online-bluegreen -Path inputs.yaml -Iac terraform -Cicd github See the cost estimates and the prerequisites.https://pgabriel-01.github.io/mlops-v3/scenarios/07-feature-store/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/07-feature-store/index.htmlAdds an Azure ML managed feature store workspace. The Parquet data source and a feature set are planned as a separate feature store workload pack. Id feature-store Network public Serving online Adds Managed feature store workspace IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Adds a managed feature store workspace alongside the project workspace. The feature store always gets a managed network in AllowInternetOutbound mode, because serverless Spark needs one and Microsoft documents it as the only supported mode for feature stores (Tutorial 6). The workspace sets the Spark runtime to 3.4.0 and has no container registry, because a managed network requires a Premium registry and the feature store does not need one.https://pgabriel-01.github.io/mlops-v3/scenarios/08-registry-promotion/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/08-registry-promotion/index.htmlAn Azure ML registry shared across environments. Models are registered in dev and promoted to test and prod. Id registry-promotion Network public Serving online Adds Shared Azure ML registry in the first environment, ml-promote workflow, deploy from the registry IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps A shared Azure ML registry is created in the first environment. ml-promote copies the model through it and deployments pull from the registry.https://pgabriel-01.github.io/mlops-v3/scenarios/09-governance/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/09-governance/index.htmlDiagnostic settings, budget alerts, resource locks and Azure Policy assignments around the workspace. Id governance Network public Serving online Adds Diagnostics, budget alerts (budget_amount, alert_emails), resource lock, policy IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Adds diagnostics to Log Analytics, a monthly budget with email alerts, a resource lock and a policy assignment per environment. Use it New-MLOpsInputFile -Scenario governance -Path inputs.yaml -Iac terraform -Cicd github Inputs to review budget_amount: see Options alert_emails: see Options See the cost estimates and the prerequisites.https://pgabriel-01.github.io/mlops-v3/scenarios/10-genai-foundry/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/10-genai-foundry/index.htmlHub workspace, AI Services account and a project for prompt flow and GenAIOps. Id genai-foundry Network public Serving none Adds AI Foundry hub, AI Services account and project IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Provisions an AI Foundry hub, an AI Services account and a project. The sample training workload is not used for GenAI; this scenario is platform only. Use it New-MLOpsInputFile -Scenario genai-foundry -Path inputs.yaml -Iac terraform -Cicd github See the cost estimates and the prerequisites.https://pgabriel-01.github.io/mlops-v3/scenarios/11-existing-platform/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/11-existing-platform/index.htmlReuses a supplied subnet, private DNS zones and Log Analytics workspace instead of creating them. Id existing-platform Network existing subnet and DNS Serving online Adds Reuses existing_platform subnet, private DNS zones and Log Analytics; governance on IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps The accelerator creates no network resources. The workspace and its private endpoints use the subnet, private DNS zones and Log Analytics workspace your platform team provides.https://pgabriel-01.github.io/mlops-v3/scenarios/12-rag-llmops/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://pgabriel-01.github.io/mlops-v3/scenarios/12-rag-llmops/index.htmlAI Foundry plus Azure AI Search, a chat and an embedding model, a sample document index, and an evaluation quality gate in the pipeline. Id rag-llmops Network public Serving none Adds Azure AI Search, chat and embedding model deployments, the aisearch Foundry connection, a rag/ workload and a rag-evaluate pipeline IaC and CI/CD Terraform or Bicep; GitHub or Azure DevOps Builds on scenario 10. Each environment gets an Azure AI Search service, a chat model and an embedding model deployed on the AI Services account, and an aisearch connection on the Foundry project. Local authentication is disabled; access uses managed identities and roles.