local-ai-models-docker-azure

SA•View on GitHub•March 9, 2025
azure
bash
docker
virtual-machine

Azure VM with AI Model Deployment

This project automates the process of spinning up a Virtual Machine (VM) on Azure, deploying AI models inside Docker containers, and accessing a web interface to interact with the models using Ollama WebUI.

Project Structure

├── azure/
│   ├── configure-network.sh       # Configures network settings for the VM
│   ├── create-resource-group.sh   # Creates a resource group in Azure
│   ├── create-vm.sh               # Creates a Virtual Machine in Azure
│   └── get-vm-details.sh          # Retrieves public IP of the created VM
├── scripts/
│   ├── deploy-containers.sh       # Deploys Docker containers on the VM
│   ├── run-model.sh               # Runs the AI model inside the container
│   └── setup-vm.sh                # Sets up the VM (install docker) and all
├── .dockerignore
├── .env
├── .env.example
├── .gitignore
├── .vm_details.env                # Stores VM public IP
├── docker-compose.yaml            # Docker Compose configuration for containers
└── README.md

Prerequisites

  • Azure CLI installed and configured.
  • Docker and Docker Compose installed on your local machine.
  • SSH access to the VM.

Usage

Step 1: Azure VM Setup

  1. Create a Resource Group Run the following script to create a resource group in Azure:

    bash azure/create-resource-group.sh

  2. Create a Virtual Machine Use this script to create a VM in the resource group:

    bash azure/create-vm.sh

  3. Retrieve VM Details After creating the VM, retrieve its details (the public IP):

    bash azure/get-vm-details.sh

  4. Configure Network Limit the connection to the user public IP

    bash azure/configure-network.sh

Step 2: VM Setup and Deployment

  1. Set Up the VM SSH into the VM and run the setup script:

    bash scripts/setup-vm.sh

  2. Deploy Docker Containers Deploy the necessary Docker containers on the VM:

    bash scripts/deploy-containers.sh

  3. Run the AI Model Start the AI model inside the Docker container:

    bash scripts/run-model.sh

Environment Variables

  • Update the .env file with your Azure credentials and other necessary configurations.
  • Use .env.example as a template for the .env file.

Notes

  • Ensure that the .vm_details.env file is generated after running get-vm-details.sh as it contains the public IP which other scripts use.

License

This project is licensed under the MIT License. See the LICENSE file for details.

GitHub