Follow Datanami:
September 28, 2021

OctoML and AMD Collaborate to Speed Up ML Deployments for Accelerated Computing

SEATTLESept. 28, 2021 — OctoML today announced a collaboration with AMD in support of Apache TVM. Now, customers, users, and communities can deploy next-generation ML applications easily on AMD CPUs, GPUs, and APUs to run almost anywhere from the cloud to the edge—including servers, laptops, gaming PCs, and more.

The OctoML and AMD collaboration, and support for AMD devices across the Apache TVM stack, offers users a uniform interface for a wide range of heterogeneous deployment hardware, and an automated process to optimize models to run with peak efficiency and optimal power.

“We are excited to collaborate with AMD as we share a common vision of a unified ML software stack across a wide array of hardware targets, enabling seamless AI/ML innovation via portability and high performance of models,” said Luis Ceze, CEO of OctoML. “This collaboration is a great example of how the industry can work successfully with the efforts of the open software community. And our customers will benefit from the ability to apply the automation capabilities of OctoML on AMD hardware.”

Built on the Apache TVM open-source project, OctoML’s commercial platform provides an automation framework that optimizes trained models to achieve optimal performance across a breadth of hardware endpoints and cloud services—without compromising accuracy. The platform is a turnkey solution that provides automated ML model performance optimization, benchmarking and packaging that accelerates the time to production.

“Deploying ML models efficiently can be difficult across multiple platforms, so users will benefit significantly from a common unified software stack to deploy their models at scale,” said Wilson Yu, Senior Director, Advanced Technologies Group, AMD. “We’re working with OctoML to help our customers create innovative ML models with the confidence that these solutions can seamlessly be deployed across supported AMD hardware.”

Visit here for more information on how AMD and OctoML collaborate to accelerate ML models.

Visit here to register for OctoML’s early access program.

About OctoML

OctoML is a machine learning acceleration platform based in Seattle, Washington. OctoML aims to accelerate model performance while enabling seamless deployment of models across any hardware platform, cloud provider, or edge device. The company’s investors include Addition, Madrona Venture Group, and Amplify Partners. OctoML was founded by creators of open-source Apache TVM, CEO Luis Ceze, CTO Tianqi Chen, CPO Jason Knight, Chief Architect Jared Roesch, and VP of Technology Partnerships Thierry Moreau. For more information, please visit

Source: OctoML