AAI 2026: AMD Launches AMD Helios Rackscale Solution for Frontier AI

New AI infrastructure combines leadership compute, networking and open software, delivering rack-scale performance for at-scale AI.

Published
AMD Helios 02 wCaption 16x9
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.

What’s the News? Today at Advancing AI 2026, AMD launched the AMD Helios™ rackscale solution, a new rackscale AI infrastructure designed for frontier AI, large-scale inference and foundation-model training. AMD Helios combines AMD Instinct™ MI455X GPUs, 6th Gen AMD EPYC™ server CPUs, AMD ROCm™ software and AMD Pensando™ networking into a unified rack-scale architecture that delivers exaflop-class AI performance, industry-leading memory capacity and bandwidth,1 and exceptional scalability.

Why It Matters? As demand grows for AI factories and frontier AI, customers need infrastructure that scales efficiently from a single rack to large gigawatt-scale AI clusters while maximizing performance, performance per watt and cost per token. AMD Helios delivers an integrated rack-scale building block that simplifies deployment while providing the compute, networking and open software foundation needed for the next generation of AI infrastructure. AMD Helios rackscale AI solution offers the highest token throughput performance at low, medium and high interactivity levels,2 designed with an open architecture from the start. Compared to the leading competitive solution, AMD Helios delivers 15% more peak FP4 performance,3 50% more high-bandwidth memory (HBM) capacity,1 6% more HBM bandwidth1 and 50% more scale-out bandwidth.4 It delivers up to 30% more tokens per dollar than the competitive solution, giving customers more output with every rack they deploy.5

What’s the Role for AMD?

Frontier AI infrastructure is evolving rapidly, and customers need platforms that can deliver exceptional performance while scaling efficiently as workloads grow. AMD Helios brings together leadership compute, high-performance networking and open software in a unified rack-scale platform. Together, these technologies give customers the flexibility to accelerate large-scale inference, frontier-model training and the next generation of AI infrastructure.
— Vamsi Boppana, senior vice president, AI, AMD

What is the AMD Helios Rackscale Solution? AMD Helios extends the AMD AI portfolio by delivering an open rack-scale architecture designed to support large-scale inference, frontier-model training and fine-tuning from a single rack to multirack clusters. AMD delivers AMD Helios at the same time AI is reshaping modern compute infrastructure, requiring tightly integrated platforms that combine high-performance compute, rack-scale architectures and open software.

  •  Built for Frontier AI: AMD Helios is the AMD rack-scale AI infrastructure platform designed for frontier AI and AI factory deployments. It provides a scalable foundation for high-volume inference, frontier-model training and fine-tuning by bringing together AMD 5th Gen Instinct GPUs, 6th Gen AMD EPYC server CPUs, AMD Pensando networking and AMD ROCm software into a unified platform built on open standards.

  • Leadership Rack Performance: AMD Helios delivers exaflop-class AI performance in a unified architecture optimized for large-scale inference and frontier-model development. A single rack provides up to 2.9 exaflops peak FP4, 1.4 exaflops peak FP8, 31 terabytes of HBM4 memory and 1.7 petabytes/second of memory bandwidth.

  • Scalable AI: AMD Helios is designed to scale from a single rack to gigawatt-scale AI clusters. Each rack integrates 18 open rack wide-aligned 4-GPU compute trays (72 total GPUs) that provide consistent building blocks for cluster deployment and expansion. At cluster scale, AMD Pensando™ Vulcano 800 AI NICs provide the high-bandwidth, low-latency connectivity required to support distributed inference, frontier-model training and rapid hyperscale growth.

  • Open Infrastructure: AMD Helios combines UALink™ over Ethernet (UALoE) for high-performance scale-up with standards-based Ethernet aligned with the Ultra Ethernet Consortium for scale-out, enabling customers to build AI infrastructure using open, standards-based technologies. The platform also incorporates defense-in-depth rack-scale security features. AMD ROCm™ software enables high-throughput inference and unified distributed training across rack and cluster-scale deployments.

More: Advancing AI 2026 (Press Kit)

Terms Expand and agree to the Terms & Conditions before downloading. Expand

Please read through and acknowledge the Terms & Conditions

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.

  • AMD Helios 02 wCaptionDownload

    The AMD Helios rackscale solution delivers an open rack-scale architecture designed to support large-scale inference, frontier-model training and fine-tuning from a single rack to multirack clusters. AMD Helios was launched at Advancing AI 2026. (Credit: AMD)

  • AMD Helios 01 wCaptionDownload

    The AMD Helios rackscale solution delivers an open rack-scale architecture designed to support large-scale inference, frontier-model training and fine-tuning from a single rack to multirack clusters. AMD Helios was launched at Advancing AI 2026. (Credit: AMD)

  • AMD Helios 03 wCaptionDownload

    The AMD Helios rackscale solution delivers an open rack-scale architecture designed to support large-scale inference, frontier-model training and fine-tuning from a single rack to multirack clusters. AMD Helios was launched at Advancing AI 2026. (Credit: AMD)

AMD Helios3 imagesThe AMD Helios rackscale solution delivers an open rack-scale architecture designed to support large-scale inference, frontier-model training and fine-tuning from a single rack to multirack clusters. AMD Helios was launched at Advancing AI 2026.Download all (3)

Cautionary Statement

This blog may contain forward-looking statements concerning Advanced Micro Devices, Inc. (AMD), which are made pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995. Forward-looking statements are commonly identified by words such as "would," "may," "expects," "believes," "plans," "intends," "projects" and other terms with similar meaning. Investors are cautioned that any forward-looking statements in this blog are based on current beliefs, assumptions and expectations, speak only as of the date of this blog and involve risks and uncertainties that could cause actual results to differ materially from current expectations. Such statements are subject to certain known and unknown risks and uncertainties, many of which are difficult to predict and generally beyond AMD's control, that could cause actual results and other future events to differ materially from those expressed in, or implied or projected by, the forward-looking information and statements. Investors are urged to review in detail the risks and uncertainties in AMD’s Securities and Exchange Commission filings, including but not limited to AMD’s most recent reports on Forms 10-K and 10-Q.  

AMD does not assume, and hereby disclaims, any obligation to update forward-looking statements made in this blog, except as may be required by law.

  1. Calculations by AMD Performance Labs in June 2026 based on published memory capacity and memory bandwidth specifications of an AMD Helios rackscale solution vs. the published memory capacity and memory bandwidth specifications of a NVIDIA Vera Rubin NVL72 rack. System manufacturers may vary configurations, yielding different results. MI400-007

  2. Based on modelling by AMD Performance Labs in July 2026, on an AMD Helios rackscale solution to measure token throughput/GPU at high, medium and low interactivity points using Kimi K2 Thinking with an ISL/OSL combination of 32K/8K compared to the published modeled specifications for an NVIDIA Vera Rubin NVL72 rack. System manufacturers may vary configurations, yielding different results. MI400-023

  3. Based on calculations by AMD Performance Labs in June 2026, to determine the peak theoretical precision performance of an AMD Instinct™ MI455X GPU using the peak Matrix FP16, BF16, INT8 datatypes and peak Open Compute Project MXFP6, MXFP8, FP8 and MXFP4 datatypes vs. published specifications of NVIDIA Vera Rubin GPU with the NVFP4 Dense datatype. System manufacturers may vary configurations, yielding different results. MI400-003

  4. Calculations by AMD Performance Labs in July 2026, based on published scale-out bandwidth of the AMD rackscale solution vs. published NVIDIA Vera Rubin NVL72 Rack specs. System manufacturers may vary configurations, yielding different results.

  5. Based on AMD Performance Labs estimates as of July 2026, tokens-per-dollar performance was calculated using the Kimi K2 Thinking workload (32K input / 8K output) on an AMD Helios rackscale solution compared to an NVIDIA Vera Rubin NVL72 rack. Results reflect estimated aggregate throughput across low, medium, and high-interactivity operating points and hourly pricing projection of system GPUs based on market conditions. System configurations may vary by manufacturer and may produce different results. MI400-025

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

Press inquiries: corporate.pressinquiry@amd.com

More from this category