Apache TVM is a high-performance open-source machine learning compiler stack designed to optimize, compile, and deploy AI models across a wide range of hardware platforms. It enables developers to improve the performance of deep learning workloads while reducing deployment complexity for production environments.

This virtual machine comes pre-installed with Apache TVM 0.23.0 on Ubuntu 26.04 LTS, providing a ready-to-use environment for AI engineers, machine learning developers, researchers, and enterprises. The image includes Python 3.14, LLVM support, essential build tools, and all required dependencies, allowing users to begin developing and optimizing machine learning applications immediately without manual installation.

Apache TVM supports model compilation and optimization for popular machine learning frameworks such as TensorFlow, PyTorch, ONNX, and Apache MXNet. It provides automatic operator tuning, graph optimization, code generation, and hardware-specific optimizations to maximize inference performance across CPUs, GPUs, and other supported accelerators.

This image is configured with Ubuntu 26.04 LTS and the latest available security updates, providing a secure, reliable, and production-ready platform for AI development, research, model optimization, and deployment. Developers can integrate TVM into existing machine learning pipelines while benefiting from its flexible compiler architecture and extensive hardware support.

Key Features

• Apache TVM 0.23.0 pre-installed and configured
• Ubuntu 26.04 LTS production-ready environment
• Python 3.14 with required TVM dependencies
• LLVM-enabled compilation support
• Support for TensorFlow, PyTorch, ONNX, and Apache MXNet models
• Machine learning model optimization and compilation
• Graph optimization and automatic operator tuning
• Ready for AI development, inference, and deployment workloads

To subscribe to this product from Azure Marketplace and deploy an instance using Azure Compute services, follow the steps below:

1. Navigate to Azure Marketplace and subscribe to the required product.
2. Search for Virtual Machines under Azure Services.
3. Click Add to open the Create a Virtual Machine page.

Create Virtual Machine

4. Under the Basics tab:
   
   • Select the appropriate subscription.
   • Create a new resource group (for example: myResourceGroup).
5. Under Instance Details:
   
   • Virtual Machine Name: myVM
   • Region: East US
   • Image: Ubuntu Server 24.04 LTS x64 Gen1/2
   • Keep remaining settings as default unless customization is required.
6. Under Administrator Account:
   
   • Authentication type: Password
   • Enter a username.
   • Create and confirm a strong password (use a secure combination of letters, numbers, and special characters).
7. Under Inbound Port Rules:
   
   • Select Allow selected ports
   • Enable:
     
     • SSH (22)
     • HTTP (80)
     • HTTPS (443)
8. Review the configuration and click Review + Create, then select Create to start deployment.

Deployment will take a few minutes to complete.

Connect to the Virtual Machine

1. Open the VM Overview page and click Connect.
2. Copy the public IP address of the virtual machine.
3. Open a terminal (Linux/macOS) or use PuTTY/Windows Terminal.
4. Connect using the following command:

ssh username@public-ip-address

5. When prompted, enter the password you created during VM setup to access the virtual machine.

Usage/Deployment Instructions

The complete deployment guide for this product is available in a detailed PDF document.
Please refer to the official deployment manual for step-by-step instructions, configuration details, and troubleshooting guidance.

Deployment Guide: Click Here