AMD and OpenAI: Collaborating Across Every Layer of the Stack

Published
Terms

Download image

Please read through and acknowledge the Terms & Conditions before downloading.

By downloading and incorporating the AMD Trademarks or AMD Logos and/or images provided herein (collectively, the “AMD Mark(s)”), you are accepting the Terms and Conditions and committing you and your company to the Terms and Conditions. Once the Terms and Conditions are accepted, AMD uses Cookies to enable the user to bypass this screen when downloading AMD brand assets in the future.

The AMD Marks are owned exclusively by Advanced Micro Devices, Inc. ("AMD"), protected under relevant trademark and copyright laws around the world, and are provided on an “as is” basis without warranty of any kind, express or implied, for use subject to the terms of this license (“License”). AMD hereby grants you ("Licensee" or “You” or “Your”) limited permission to use the AMD Mark(s) solely in connection with the advertising, marketing, promotion, and sale of Licensee products that include an AMD CPU, GPU, APU, or FPGA (each an “AMD Processor”) and in accordance with the AMD Corporate and Product Branding Guidelines, which may be modified by AMD from time to time. You agree to ensure that You are in possession of the latest, updated version.

This grant does not extend to use of AMD Mark(s) on Licensee product packaging or use of AMD Mark(s) on any goods that do not contain an AMD Processor. AMD shall remain the exclusive owner of all rights in the AMD Mark(s), and this License does not grant any other rights with respect to AMD’s Mark(s) trademarks or copyright. You agree that You will not, nor will you help others, contest, attack, or dispute AMD’s rights in the AMD Mark(s). The AMD Mark(s) must be used only in a positive manner and shall not be used in any way that may, in AMD’s sole judgement, disparage AMD’s reputation.

You agree not to use the AMD Mark(s) in a manner to suggest AMD’s endorsement or sponsorship of your goods or services, or in any way that may create confusion as to the source of the goods, and Licensee’s company name, logo or product name shall appear on any products or related materials where an AMD Mark(s) is used. This license may be terminated by AMD at any time, and failure to comply with these terms will result in automatic termination of this License.

Sometimes AMD acquires the right(s) to certain image(s) and includes such image(s) within AMD system(s) (“Licensed Images”). You agree to refrain from using Licensed Images for any reason outside of those directly related to your reason for accessing AMD system(s) contemplated under these Terms and Conditions. Additionally, you agree to comply with all limitation(s) on Licensed Images that are communicated to you by AMD, including but not limited to limitation(s) conveyed to you via notes in a file containing the Licensed Images.

If You encounter any non-public information related to AMD (“Confidential Information”) while accessing AMD system(s), You agree to (a) reasonably protect and not disclose any Confidential Information to any third party(ies) without AMD’s written permission; and (b) refrain from using Confidential Information in ways not relating to your access of AMD system(s).

If it is determined by AMD – in its sole opinion - that You have violated these Terms and Conditions or have otherwise misused Confidential Information, Your access to AMD system(s) may be revoked immediately and without notice.

If you have any questions regarding the terms and conditions of this Agreement, please contact the AMD Law Department at amd.trademarks@amd.com.

Three things you need to know:

  • OpenAI feeds AMD early insight into where models are heading; AMD turns that into silicon, systems and software. This loop has shaped the AMD Instinct™ MI400 generation and now extends toward MI500.

  • Work spanning Triton, Gluon, LLVM and AMD ROCm™ – plus Codex and agentic tools that help write GPU code – means improvements don’t stay locked to one workload; they strengthen the broader AMD ecosystem developers can build on.

  • OpenAI expects to bring AMD Helios™ online through multiple deployment partners starting in the second half of 2026, marking the initial phase of the 6-gigawatt AMD GPU deployment the companies announced in October 2025.

 At Advancing AI 2026, AMD and OpenAI showcased a partnership evolving from product collaboration to full-stack co-design. AMD Chair and CEO Dr. Lisa Su and OpenAI Vice President of Compute Strategy Sachin Katti discussed rising compute demand, the work underway to bring AMD Helios™ online, and the need to shape future infrastructure around emerging model requirements.

AMD’s Vamsi Boppana and OpenAI’s Philippe Tillet then turned to the software behind that progress – from Triton and Gluon to compiler optimization and AI-assisted GPU programming. Together, the discussions revealed how close collaboration can turn advances in silicon and software into deployable performance for frontier AI systems.

From MI300X to Helios

Katti said AI models are becoming more capable, more agentic and useful across a broader range of tasks. As AI increasingly helps improve software, research and infrastructure itself, demand continues to rise for greater performance, memory, networking and efficiency.

The AMD and OpenAI collaboration began with AMD Instinct MI300X GPUs and deepened with MI355X as the teams optimized the software stack, networking and large-scale deployment. The AMD Helios™ rackscale solution, with AMD Instinct™ MI455X GPUs and 6th Gen AMD EPYC™ "Venice" CPUs, has extended the relationship, with engineers from AMD and OpenAI working side by side across hardware, software and systems.

OpenAI has had access to Helios systems for several months, allowing its teams to prepare the software and solution stack ahead of deployment. The companies are optimizing GPT-class workloads on the platform, debugging together, sharing data and making decisions quickly.

OpenAI expects to begin bringing Helios online through multiple deployment partners in the second half of 2026, with deployments accelerating throughout 2027. This rollout will mark the first phase of the 6-gigawatt deployment the companies announced in October 2025.

Shaping Future Infrastructure Together

The collaboration extends beyond preparing today’s systems for deployment. AMD and OpenAI are aligning their roadmaps so future infrastructure is shaped by the workloads it will serve.

At the frontier, performance depends on how compute, memory, networking, power, cooling, rack architecture, software and models operate together. OpenAI gives AMD early insight into how models are changing, where bottlenecks are emerging and what future training and inference workloads will require. AMD can translate those insights into advances across silicon, systems and software.

That feedback loop has helped shape the AMD Instinct MI400 generation and is continuing as the companies look further ahead. Katti said OpenAI is excited by what AMD is designing with the Instinct MI500 generation, which looks set to deliver another significant leap in AI capabilities.

Turning Hardware Advances into Application Performance

Tillet emphasized that there is no one-size-fits-all approach to GPU software. OpenAI chooses the programming model that best fits each workload.

For research, where speed of iteration is critical, Triton remains a powerful tool for developing GPU kernels quickly. For performance-sensitive workloads such as inference, OpenAI has developed Gluon, a more hardware-aware programming language that gives engineers greater control over how software uses the GPU.

AMD and OpenAI are collaborating across both. Their work now spans the software stack down to LLVM code generation, ensuring hardware advances translate into meaningful application performance.

Codex is also becoming an increasingly valuable part of this process. It is already strong at helping engineers write Triton code and is becoming more capable with Gluon, allowing teams to explore implementations and adapt software to new hardware more rapidly.

Using AI to Build Better AI Infrastructure

AMD and OpenAI are also jointly researching how AI can improve the systems on which AI runs.

Today, expert engineers invest substantial effort translating new models into efficient kernels, compilers, communication libraries and system configurations. AMD and OpenAI are developing technologies that could allow models and agents to automate more of that work

The conversations at Advancing AI 2026 made clear that the AMD and OpenAI relationship is about more than deploying infrastructure at scale. By connecting model development with silicon, systems and software, the companies are creating a faster and more collaborative approach to building the infrastructure behind the next generation of AI.

 

TopicsArtificial IntelligenceData Center #rocm #epyc #instinct #2026-aai #helios
Share / cite

Press inquiries: corporate.pressinquiry@amd.com

More from this category