AMD GPU Infrastructure Donation

Through the AMD-U of T Research Lab, AMD has made a significant contribution to the University of Toronto’s AI and high-performance computing infrastructure, donating:

  • Two 8-way AMD Instinct™ MI350X GPU systems
  • One 8-way AMD Instinct™ MI100 GPU system

 

The AMD Instinct™ MI350X is AMD’s flagship AI accelerator, built on the company’s latest 4th Generation CDNA architecture.

Each MI350X GPU features 288 GB of high-speed HBM3E memory, providing researchers with access to cutting-edge computing resources for large-scale artificial intelligence, machine learning, data science, and high-performance computing workloads.

Following Installation and Configuration

AMD engineers partnered with the University of Toronto to deliver hands-on training sessions covering the AMD software ecosystem, including ROCm™ and HIP, enabling researchers to efficiently develop and optimize applications on AMD GPU platforms.

Now these systems are actively supporting research across the Department of Computer Science and the broader University of Toronto community. Faculty members, postdoctoral researchers, and students are using the infrastructure to advance work in areas including foundation models, generative AI, scientific computing, advanced data analytics, and other data-intensive research domains.

The donated systems represent a cornerstone of the AMD-U of T Research Lab’s commitment to expanding access to world-class AI computing resources and strengthening industry-academic collaboration. Through this investment, AMD is helping accelerate research innovation while supporting the training and development of the next generation of AI talent at the University of Toronto.

AMD Training
AMD Training

Contact

Learn more about the partnerships, programs and opportunities.

Wendy Wu

Research Partnerships Lab Manager
[email protected]