MEMORY INDUSTRY INTELLIGENCE
AAI 2026: AMD, 에이전틱 AI 시대를 위한 풀스택 컴퓨트 제공
한국어 번역·요약·분석
원문 제목: AAI 2026: AMD Delivers Full-Stack Compute for the Agentic AI Era
원문 문장이 일치하지 않은 주장 2건은 근거 등록에서 제외했습니다.
핵심 요약
AMD는 Advancing AI 2026에서 6세대 EPYC CPU, Instinct MI400 시리즈 GPU, Helios 랙스케일 AI 솔루션, Ryzen AI Embedded X100 프로세서, Kria AI SOM 및 로보틱스 개발자 플랫폼을 발표했다. Helios는 72개의 MI455X GPU와 18개의 6세대 EPYC 'Venice' CPU를 통합해 경쟁 솔루션 대비 달러당 최대 30% 더 많은 추론 토큰을 제공한다고 주장했다. Anthropic은 최대 2GW 규모의 MI455X GPU를 Helios 랙에 배치하기로 했고, OpenAI는 2026년 4분기부터 Helios를 온라인에 올릴 예정이며, Meta는 6세대 EPYC 플랫폼과 Helios 랙을 검증 중이다. AMD는 2027년 MI500, 2028년 MI600 GPU와 Zen 7 기반 EPYC, 2030년 Zen 8 기반 'Ravenna' CPU 등 로드맵을 공개했다. 이는 AI 인프라 확장에 따른 서버 CPU·GPU 수요 증가를 보여주지만, 메모리 산업에 대한 직접적 영향은 원문에서 명시적으로 확인되지 않는다.
메모리 산업 영향 분석
AMD의 Helios 랙스케일 솔루션과 6세대 EPYC CPU는 서버 DRAM 및 HBM 수요에 긍정적일 수 있다. Helios는 72개의 MI455X GPU와 18개의 EPYC CPU를 통합하며, EPYC 6세대는 최대 16채널 12.8 GT/s MRDIMM을 지원한다고 명시되어 있어 서버 DRAM 채널당 모듈 수요 증가 가능성이 있다. 그러나 원문은 메모리 공급업체나 HBM 탑재량, 서버 DRAM 용량을 구체적으로 언급하지 않아 메모리 산업에 대한 직접적 영향은 확인되지 않는다. Anthropic의 2GW 배치, OpenAI의 2026년 4분기 Helios 온라인 계획은 GPU 출하량 증가를 시사하지만, 이에 따른 HBM 수요 증가는 분석가 가설이다. NAND/eSSD와의 직접 연결 근거는 원문에 없다. 확인할 지표로는 MI455X GPU당 HBM 스택 수, EPYC 6세대 서버의 DRAM 탑재량, Helios 랙 출하량이 있다.
한국어 번역 읽기
수집된 원문 v1의 분석에 제공된 본문 기준 · 14960자
AMD는 첫 랙스케일 AI 솔루션이자 세계에서 가장 강력한 AI 랙인 AMD Helios를 포함한 광범위한 고성능 컴퓨팅 및 피지컬 AI 포트폴리오를 출시한다.
**뉴스 하이라이트**
* Advancing AI 2026에서 AMD는 6세대 AMD EPYC™ CPU, AMD Instinct™ MI400 시리즈 GPU, AMD Helios™ AI 랙스케일 솔루션, AMD Ryzen™ AI Embedded X100 프로세서, AMD Kria™ AI SOM 및 로보틱스 개발자 플랫폼을 출시한다.
* AMD Helios는 경쟁 제품보다 달러당 최대 30% 더 많은 추론 토큰을 제공하여 배치된 모든 랙에서 출력을 극대화한다.
* AI는 데이터센터, PC, 엣지 및 임베디드 프로세서에 걸친 AMD 실리콘 전반에 대한 수요를 가속화하여 AMD의 TAM을 2030년까지 약 2조 달러로 끌어올린다.
* Anthropic, OpenAI, Meta, Cerebras, AT&T, Cisco는 AMD와 협력하여 AI 인프라, 엔터프라이즈 기능 및 엣지 AI를 발전시키는 방법을 상세히 설명했다.
샌프란시스코, 2026년 7월 23일 (GLOBE NEWSWIRE) -- [AMD](http://www.amd.com/ "새 창에서 열림") (NASDAQ: AMD)는 오늘 Advancing AI 2026에서 차세대 AI 인프라 및 피지컬 AI 포트폴리오를 출시했으며, 이는 현재 생산 중이어서 선도적인 AI 기업들이 기가와트 규모로 배치할 AMD Helios 랙스케일 솔루션이 주도한다.
AI가 훈련에서 추론 및 에이전틱 워크로드로 확장됨에 따라 컴퓨트 수요가 빠르게 가속화되고 있다. AMD는 고객이 모든 워크로드에 적합한 컴퓨트를 배치할 수 있는 유연성을 제공하는 개방형 풀스택 AI 플랫폼을 제공한다.
“AI의 다음 단계는 프론티어 모델, 에이전트 및 피지컬 AI에 걸쳐 있으며, 지능을 모든 곳에 가져올 새로운 기회를 창출할 것입니다.”라고 AMD 회장 겸 CEO인 리사 수 박사는 말했다. “그 잠재력을 실현하려면 전체 산업이 함께 노력해야 합니다. AMD는 생태계 전반에 걸쳐 파트너십을 맺어 고객이 데이터센터에서 엣지까지 AI를 확장할 수 있는 성능, 유연성 및 선택권을 제공하는 리더십 컴퓨트와 개방형 플랫폼을 제공합니다.”
**AMD Helios: 최고 성능의 랙스케일 AI 솔루션**
프론티어 AI를 제공하려면 스택의 모든 부분이 성능의 한계를 밀어붙이는 완전히 통합된 랙 아키텍처가 필요하다. [AMD Helios 랙스케일 솔루션](https://newsroom.amd.com/news/aai-2026-helios-update "새 창에서 열림")은 이를 위해 구축되었으며, 72개의 고성능 AMD Instinct™ MI455X GPU와 18개의 강력한 6세대 AMD EPYC™ “Venice” CPU에 걸친 공동 최적화 실리콘, AMD Pensando™ 프런트엔드, 스케일업 및 스케일아웃 네트워킹으로 연결되고 AMD ROCm™ 개방형 소프트웨어로 가속된다. AMD Helios는 리더십 컴퓨트 성능, 메모리 용량 및 네트워킹 대역폭을 결합하여 선도적인 경쟁 솔루션보다 달러당 최대 30% 더 많은 토큰을 제공한다 1.
선도적인 AI 연구소와 클라우드 제공업체는 개방형 풀스택 성능을 위해 AMD Helios를 선택하고 있다. 여기에는 OpenAI, Anthropic, Meta, Microsoft, Oracle, HUMAIN, Tensorwave, Vultr, Cirrascale 등이 포함된다. 시스템은 Bull, HPE, Lenovo, Supermicro를 포함한 선도적인 OEM과 인프라 파트너 Sanmina 및 Wiwynn에서 제공될 예정이다.
Advancing AI에서 AMD 파트너들은 프론티어 훈련 및 추론을 위해 AMD AI 인프라를 대규모로 배치하는 방법을 상세히 설명했다:
* Anthropic과 AMD는 수요일의 전략적 파트너십 발표를 추가로 설명하며 AMD Helios 랙스케일 솔루션에 최대 2기가와트의 AMD Instinct MI455X GPU를 배치하기로 했다. 두 회사는 Claude를 사용하여 AMD 소프트웨어 개발을 가속화하는 다년간 엔지니어링 협력을 시작한다. 구체적으로, 팀은 Claude를 사용하여 AMD Instinct GPU용 워크로드를 최적화하고 ROCm 소프트웨어 개발을 가속화할 것이다. AMD는 또한 엔지니어링 및 제품 개발 팀 전반에 Claude를 광범위하게 채택할 것이다.
* OpenAI와 AMD는 실리콘에서 소프트웨어까지 전체 AI 스택을 최적화하기 위해 파트너십을 맺고 있다. OpenAI의 Triton 프레임워크와 AMD ROCm 소프트웨어를 활용하여 두 회사는 AMD Instinct MI455X GPU 및 AMD Helios 랙에서 GPT급 워크로드를 최적화하고 있다. OpenAI는 2026년 4분기부터 Helios를 온라인에 올릴 예정이며, 2027년 내내 배치가 가속화될 것으로 예상한다.
* Meta와 AMD는 기가와트 규모 배치를 위해 공동 설계하고 있으며, Meta 워크로드에 맞게 AMD의 전체 AI 컴퓨트 스택을 최적화하고 있다. Meta는 현재 자사 연구소에서 6세대 EPYC CPU 플랫폼을 검증하고 있으며, 대규모 배치를 준비하면서 AMD Helios 랙에서 워크로드를 테스트하고 검증하기 시작했다.
* Cerebras와 AMD는 [협력](https://newsroom.amd.com/news/aai-2026-cerebras-inference "새 창에서 열림")하여 Cerebras의 초저지연 AI 컴퓨트와 AMD Helios의 고처리량 랙스케일 인프라를 결합한 솔루션을 제공하여 초저지연 추론 서빙의 추론 효율성, 확장성 및 경제성을 개선하는 데 도움을 준다.
**최고 성능의 데이터센터 CPU 및 GPU 제공**
6세대 EPYC 프로세서는 에이전틱 AI를 위한 가장 광범위한 서버 CPU 포트폴리오를 제공하며,2 클라우드, 엔터프라이즈, 범용 및 고성능 컴퓨팅(HPC) 워크로드에 걸쳐 있다. 리더십 코어당 성능과 최고 스레드 밀도3를 통해 와트당, 달러당, 랙당 가장 많은 에이전트를 가능하게 한다.4,5,6 AI 호스트 노드의 경우, [6세대 EPYC CPU](https://newsroom.amd.com/news/aai-2026-6th-gen-epyc "새 창에서 열림")는 가속기를 완전히 공급할 수 있는 속도와 메모리 대역폭을 제공한다. 그리고 범용 서버의 경우, 비즈니스 크리티컬 애플리케이션과 AI 지원 작업을 실행하는 리더십 성능과 에너지 효율성을 제공한다.
[AMD Instinct™ MI400 시리즈 GPU](https://newsroom.amd.com/news/aai-2026-mi400-instinct-update "새 창에서 열림")를 통해 AMD는 클라우드, 엔터프라이즈 및 HPC 워크로드에 강력한 성능을 제공한다. AMD Instinct MI455X GPU는 MI355X GPU 대비 34배 높은 토큰 처리량을 제공한다 7. 고정밀 워크로드의 경우, AMD Instinct™ MI430X 가속기는 과학 컴퓨팅을 위한 최대 288 TFLOPS의 하드웨어 기반 FP64 성능을 갖춘 HPC 및 소버린 AI를 위한 가장 진보된 제품이다. Instinct MI430X 가속기는 미국과 유럽 전역에서 차세대 엑사스케일급 슈퍼컴퓨터를 구동하고 있다.
AMD는 또한 기존 인프라에 원활한 AI 가속을 제공하는 Instinct MI350P GPU를 출시하여 리더십 토큰 경제성을 제공한다. MI350P GPU는 경쟁 제품보다 달러당 최대 4.2배 더 많은 초당 토큰을 제공한다 8.
**개방형 소프트웨어 생태계 발전**
개발자를 위해 AMD ROCm은 AMD 하드웨어에서 AI를 구축하고 배포하는 데 필요한 성능, 유연성 및 생태계 지원을 갖춘 개방형 소프트웨어 플랫폼이다. 그 기반 위에서 AMD는 [ROCm.ai](https://newsroom.amd.com/news/aai-2026-rocm-ai-software "새 창에서 열림")를 도입한다. 이는 개발자가 AMD 플랫폼 전반에서 GPU 소프트웨어를 더 빠르게 구축, 최적화 및 배포할 수 있도록 돕는 AI 기반 개발 플랫폼이다. ROCm.ai는 Claude, Codex, Cursor와 같은 인기 있는 코딩 에이전트가 AMD 플랫폼과 ROCm을 기본적으로 이해할 수 있게 함으로써 AI 지원 GPU 프로그래밍을 개발자에게 제공한다.
ROCm.ai는 AMD Instinct MI455X GPU의 소프트웨어 지원을 가속화하는 동시에 성능을 최적화하고 있다. PyTorch, Hugging Face, vLLM, SGLang을 포함한 선도적인 오픈소스 프레임워크는 이미 MI455X에서 지원되며 좋은 결과를 보이고 있다.
**차세대 AI 인프라 가속화**
AMD는 2030년까지 CPU, GPU, 네트워킹 및 랙스케일 혁신의 연간 케이던스를 확장하고 있다. 회사는 로드맵에 대한 새로운 세부 정보를 공유했으며, 여기에는 다음이 포함된다:
* “Zen 7” 아키텍처 기반 차세대 EPYC 서버 CPU는 2028년에 출시된다. “Florence”, “Ferrara”, “Fidenza” CPU는 밀도, 성능, 시스템 달러당 성능 및 와트당 성능에서 AMD의 리더십을 확장할 것으로 예상된다.
* “Zen 8” 아키텍처 기반 “Ravenna” CPU는 2030년에 출시되어 AMD 서버 CPU 리더십을 지속하도록 설계되었다.
* 차세대 AMD Instinct MI500 시리즈 GPU는 2027년에 출시되며, 리더십 성능을 위한 차세대 컴퓨트, 메모리 및 인터커넥트 기술을 갖춘다.
* AMD Instinct MI600 시리즈 GPU는 2028년에 출시된다.
* 차세대 AMD Helios 500 랙스케일 솔루션은 AMD Instinct MI500 시리즈 GPU와 AMD EPYC “Verano” CPU, 차세대 Pensando “Como” 및 “Monza” 네트워킹으로 구동된다. AMD Helios 600 랙스케일 솔루션은 AMD Instinct MI600 시리즈 GPU, EPYC “Ferrara” CPU 및 Pensando “Palma” 및 “Levanzo” 네트워킹으로 구동되며 뒤따를 것이다.
**엔터프라이즈 전반에 AI 확장**
선도적인 엔터프라이즈는 클라우드, 하이브리드 및 온프레미스 데이터센터에서 AI 지원 PC 플릿에 이르기까지 AMD 인프라에서 실행된다. AMD 기술은 고객이 빠르게 확장하고 엔터프라이즈 전환을 가속화하도록 돕고 있다.
Advancing AI에서 AT&T는 클라우드, 온프레미스 및 에어갭 환경 전반에 걸쳐 AMD 기술을 사용하여 유연한 엔터프라이즈 AI를 배포하는 방법을 설명했다. AT&T는 또한 AMD Instinct GPU와 ROCm 소프트웨어를 사용하여 통신용으로 특별히 훈련된 오픈소스 모델인 OTel 2.0 모델을 구동하고 있다.
AMD Ryzen™ AI Halo 개발자 플랫폼을 통해 AMD는 로컬 AI 개발을 접근 가능하게 만드는 성능, 효율성 및 단순성을 제공한다. Ryzen™ AI Max PRO 400 시리즈 프로세서로 구동되는 더 많은 AMD Ryzen AI Halo 플랫폼이 올해 말 AMD 및 OEM 파트너로부터 제공될 예정이다.
Cisco와 AMD는 AMD Ryzen AI Halo 시스템을 포함한 AMD 고성능 추론 엔진을 Cisco 네트워킹, 관찰 가능성 및 보안 기능과 결합하여 기업이 하이브리드 및 로컬 에이전틱 AI를 대규모로 배포, 관리 및 거버넌스할 수 있도록 협력하고 있다.
**피지컬 AI의 차세대 프론티어 발전**
AI가 클라우드, 엔터프라이즈 및 로컬 시스템 전반으로 확장됨에 따라 다음 프론티어는 물리적 세계에서 인식, 추론 및 행동하는 기계에 지능을 가져오는 것이다. AMD FPGA 및 적응형 SoC를 통한 로보틱스의 오랜 유산을 바탕으로 AMD는 로봇 몸체에서 로봇 두뇌로 회사의 로보틱스 역량을 확장하는 AMD Kria™ AI 솔루션을 도입했다. AMD는 AI 인식, 추론 및 에이전틱 의사 결정과 제어를 단일 플랫폼에서 함께 제공하여 까다로운 실제 로봇 시스템에 필요한 성능을 제공하는 독보적인 위치에 있다.
포트폴리오에는 새로운 [AMD Kria AI 시스템 온 모듈](https://www.amd.com/en/products/system-on-modules/kria/ai.html "새 창에서 열림")(SOM)과 새로운 [AMD Ryzen AI Embedded X100 시리즈 프로세서](https://newsroom.amd.com/news/aai-2026-ryzen-ai-embedded-x100 "새 창에서 열림")로 구동되는 것, 그리고 CPU, GPU, NPU 및 FPGA 컴퓨트를 결합한 자율 로보틱스용 최초의 개방형 턴키 통합 플랫폼인 [AMD Kria AI 로보틱스 개발자 플랫폼](https://newsroom.amd.com/news/aai-2026-kria-robotics-dev-platform "새 창에서 열림")이 포함된다. 확장된 개방형 소프트웨어 생태계와 함께 AMD Kria AI 솔루션은 벤더 종속성을 제거하고 개발자와 고객이 차세대 피지컬 AI 시스템의 프로토타입에서 생산까지 가는 경로를 가속화하도록 돕는다.
**지원 리소스**
* [Advancing AI 2026](https://newsroom.amd.com/press-kits/advancing-ai-2026-all-news "새 창에서 열림")에서 AMD 팔로우 (프레스 키트)
* [AAI 기조연설 다시보기](https://newsroom.amd.com/news/aai-2026-keynote-livestre
[원문 후반 발췌]
logy or other intellectual property; unfavorable currency exchange rate fluctuations; ability of third party manufacturers to manufacture AMD's products on a timely basis in sufficient quantities and using competitive technologies; availability of essential equipment, materials, components (such as memory supply), substrates or manufacturing processes; ability to achieve expected manufacturing yields for AMD’s products; AMD's ability to generate revenue from its semi-custom SoC products; potential security vulnerabilities; potential security incidents including IT outages, data loss, data breaches and cyberattacks; uncertainties involving the ordering and shipment of AMD’s products; AMD’s reliance on third-party intellectual property to design and introduce new products; AMD's reliance on third-party companies for design, manufacture and supply of motherboards, software, memory and other computer platform components; AMD's reliance on Microsoft and other software vendors' support to design and develop software to run on AMD’s products; AMD’s reliance on third-party distributors and add-in-board partners; impact of modification or interruption of AMD’s internal business processes and information systems; compatibility of AMD’s products with some or all industry-standard software and har
[원문 후반 발췌]
general-purpose enterprise, cloud, telecom, SMB, and HPC systems to emerging AI environments including sandboxed agentic AI deployments and GPU head node servers. AMD EPYC 6th Generation platforms extend this breadth by uniquely combining high core and thread density of up to 512 threads, advanced memory bandwidth of up to 16 channels of 12.8 GT/s MRDIMM support, next-generation PCIe® Gen 6 connectivity, and select SKUs with boost frequencies up to 5 GHz.
3 EPYC-025D: As of July 2026, 6th Gen EPYC 9996 has 256 cores and 512 threads with SMT enabled which is higher than any other publicly disclosed 1P CPU
4 9xx6-012: Based on estimated performance data for Nvidia, Intel®, and AMD EPYC™ Server Processors for Agentic AI, the AMD EPYC9996 provides the most agents per rack at a 100Kw power envelope per rack.
Compared to the cores per rack of Nvidia Vera (88c) powered server racks:
- The AMD EPYC 9996 (256C) provides 2.08x the cores (and threads with SMT) per rack
- The AMD EPYC 9965 (192C) provides 1.86x the cores(and threads with SMT) per rack
- The Intel Xeon 6980P (128C) provides 1.24x the cores(and threads with SMT) per rack
Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf
Agent counts are estimates derived from available
[원문 후반 발췌]
Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
6 9xx6-014: Comparison based on published Top-of-stack core counts and CPU W, Default CPU Power for 6th Gen EPYC, and TDPs for 5th Gen EPYC, Intel® Xeon®, and Nvidia Vera for estimated Highest Agents / CPU W across 6th Gen AMD EPYC™ 256C at 400W, AMD EPYC™ 9965 (500W TDP), Nvidia Vera (450W TDP), ARM AGI (300W TDP), Intel® Xeon® 6980P (500W TDP), and Intel Xeon 6990E+ (450W TDP) powered servers as of 7/22/2026. 2 threads per core (SMT). 1 thread per core for ARM AGI.
Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
Starting with the 6th Gen AMD EPYC™ server processor family, AMD uses Default CPU Power to describe processor power consumption, succeeding AMD's historical TDP reference. Default CPU Power reflects total power
**뉴스 하이라이트**
* Advancing AI 2026에서 AMD는 6세대 AMD EPYC™ CPU, AMD Instinct™ MI400 시리즈 GPU, AMD Helios™ AI 랙스케일 솔루션, AMD Ryzen™ AI Embedded X100 프로세서, AMD Kria™ AI SOM 및 로보틱스 개발자 플랫폼을 출시한다.
* AMD Helios는 경쟁 제품보다 달러당 최대 30% 더 많은 추론 토큰을 제공하여 배치된 모든 랙에서 출력을 극대화한다.
* AI는 데이터센터, PC, 엣지 및 임베디드 프로세서에 걸친 AMD 실리콘 전반에 대한 수요를 가속화하여 AMD의 TAM을 2030년까지 약 2조 달러로 끌어올린다.
* Anthropic, OpenAI, Meta, Cerebras, AT&T, Cisco는 AMD와 협력하여 AI 인프라, 엔터프라이즈 기능 및 엣지 AI를 발전시키는 방법을 상세히 설명했다.
샌프란시스코, 2026년 7월 23일 (GLOBE NEWSWIRE) -- [AMD](http://www.amd.com/ "새 창에서 열림") (NASDAQ: AMD)는 오늘 Advancing AI 2026에서 차세대 AI 인프라 및 피지컬 AI 포트폴리오를 출시했으며, 이는 현재 생산 중이어서 선도적인 AI 기업들이 기가와트 규모로 배치할 AMD Helios 랙스케일 솔루션이 주도한다.
AI가 훈련에서 추론 및 에이전틱 워크로드로 확장됨에 따라 컴퓨트 수요가 빠르게 가속화되고 있다. AMD는 고객이 모든 워크로드에 적합한 컴퓨트를 배치할 수 있는 유연성을 제공하는 개방형 풀스택 AI 플랫폼을 제공한다.
“AI의 다음 단계는 프론티어 모델, 에이전트 및 피지컬 AI에 걸쳐 있으며, 지능을 모든 곳에 가져올 새로운 기회를 창출할 것입니다.”라고 AMD 회장 겸 CEO인 리사 수 박사는 말했다. “그 잠재력을 실현하려면 전체 산업이 함께 노력해야 합니다. AMD는 생태계 전반에 걸쳐 파트너십을 맺어 고객이 데이터센터에서 엣지까지 AI를 확장할 수 있는 성능, 유연성 및 선택권을 제공하는 리더십 컴퓨트와 개방형 플랫폼을 제공합니다.”
**AMD Helios: 최고 성능의 랙스케일 AI 솔루션**
프론티어 AI를 제공하려면 스택의 모든 부분이 성능의 한계를 밀어붙이는 완전히 통합된 랙 아키텍처가 필요하다. [AMD Helios 랙스케일 솔루션](https://newsroom.amd.com/news/aai-2026-helios-update "새 창에서 열림")은 이를 위해 구축되었으며, 72개의 고성능 AMD Instinct™ MI455X GPU와 18개의 강력한 6세대 AMD EPYC™ “Venice” CPU에 걸친 공동 최적화 실리콘, AMD Pensando™ 프런트엔드, 스케일업 및 스케일아웃 네트워킹으로 연결되고 AMD ROCm™ 개방형 소프트웨어로 가속된다. AMD Helios는 리더십 컴퓨트 성능, 메모리 용량 및 네트워킹 대역폭을 결합하여 선도적인 경쟁 솔루션보다 달러당 최대 30% 더 많은 토큰을 제공한다 1.
선도적인 AI 연구소와 클라우드 제공업체는 개방형 풀스택 성능을 위해 AMD Helios를 선택하고 있다. 여기에는 OpenAI, Anthropic, Meta, Microsoft, Oracle, HUMAIN, Tensorwave, Vultr, Cirrascale 등이 포함된다. 시스템은 Bull, HPE, Lenovo, Supermicro를 포함한 선도적인 OEM과 인프라 파트너 Sanmina 및 Wiwynn에서 제공될 예정이다.
Advancing AI에서 AMD 파트너들은 프론티어 훈련 및 추론을 위해 AMD AI 인프라를 대규모로 배치하는 방법을 상세히 설명했다:
* Anthropic과 AMD는 수요일의 전략적 파트너십 발표를 추가로 설명하며 AMD Helios 랙스케일 솔루션에 최대 2기가와트의 AMD Instinct MI455X GPU를 배치하기로 했다. 두 회사는 Claude를 사용하여 AMD 소프트웨어 개발을 가속화하는 다년간 엔지니어링 협력을 시작한다. 구체적으로, 팀은 Claude를 사용하여 AMD Instinct GPU용 워크로드를 최적화하고 ROCm 소프트웨어 개발을 가속화할 것이다. AMD는 또한 엔지니어링 및 제품 개발 팀 전반에 Claude를 광범위하게 채택할 것이다.
* OpenAI와 AMD는 실리콘에서 소프트웨어까지 전체 AI 스택을 최적화하기 위해 파트너십을 맺고 있다. OpenAI의 Triton 프레임워크와 AMD ROCm 소프트웨어를 활용하여 두 회사는 AMD Instinct MI455X GPU 및 AMD Helios 랙에서 GPT급 워크로드를 최적화하고 있다. OpenAI는 2026년 4분기부터 Helios를 온라인에 올릴 예정이며, 2027년 내내 배치가 가속화될 것으로 예상한다.
* Meta와 AMD는 기가와트 규모 배치를 위해 공동 설계하고 있으며, Meta 워크로드에 맞게 AMD의 전체 AI 컴퓨트 스택을 최적화하고 있다. Meta는 현재 자사 연구소에서 6세대 EPYC CPU 플랫폼을 검증하고 있으며, 대규모 배치를 준비하면서 AMD Helios 랙에서 워크로드를 테스트하고 검증하기 시작했다.
* Cerebras와 AMD는 [협력](https://newsroom.amd.com/news/aai-2026-cerebras-inference "새 창에서 열림")하여 Cerebras의 초저지연 AI 컴퓨트와 AMD Helios의 고처리량 랙스케일 인프라를 결합한 솔루션을 제공하여 초저지연 추론 서빙의 추론 효율성, 확장성 및 경제성을 개선하는 데 도움을 준다.
**최고 성능의 데이터센터 CPU 및 GPU 제공**
6세대 EPYC 프로세서는 에이전틱 AI를 위한 가장 광범위한 서버 CPU 포트폴리오를 제공하며,2 클라우드, 엔터프라이즈, 범용 및 고성능 컴퓨팅(HPC) 워크로드에 걸쳐 있다. 리더십 코어당 성능과 최고 스레드 밀도3를 통해 와트당, 달러당, 랙당 가장 많은 에이전트를 가능하게 한다.4,5,6 AI 호스트 노드의 경우, [6세대 EPYC CPU](https://newsroom.amd.com/news/aai-2026-6th-gen-epyc "새 창에서 열림")는 가속기를 완전히 공급할 수 있는 속도와 메모리 대역폭을 제공한다. 그리고 범용 서버의 경우, 비즈니스 크리티컬 애플리케이션과 AI 지원 작업을 실행하는 리더십 성능과 에너지 효율성을 제공한다.
[AMD Instinct™ MI400 시리즈 GPU](https://newsroom.amd.com/news/aai-2026-mi400-instinct-update "새 창에서 열림")를 통해 AMD는 클라우드, 엔터프라이즈 및 HPC 워크로드에 강력한 성능을 제공한다. AMD Instinct MI455X GPU는 MI355X GPU 대비 34배 높은 토큰 처리량을 제공한다 7. 고정밀 워크로드의 경우, AMD Instinct™ MI430X 가속기는 과학 컴퓨팅을 위한 최대 288 TFLOPS의 하드웨어 기반 FP64 성능을 갖춘 HPC 및 소버린 AI를 위한 가장 진보된 제품이다. Instinct MI430X 가속기는 미국과 유럽 전역에서 차세대 엑사스케일급 슈퍼컴퓨터를 구동하고 있다.
AMD는 또한 기존 인프라에 원활한 AI 가속을 제공하는 Instinct MI350P GPU를 출시하여 리더십 토큰 경제성을 제공한다. MI350P GPU는 경쟁 제품보다 달러당 최대 4.2배 더 많은 초당 토큰을 제공한다 8.
**개방형 소프트웨어 생태계 발전**
개발자를 위해 AMD ROCm은 AMD 하드웨어에서 AI를 구축하고 배포하는 데 필요한 성능, 유연성 및 생태계 지원을 갖춘 개방형 소프트웨어 플랫폼이다. 그 기반 위에서 AMD는 [ROCm.ai](https://newsroom.amd.com/news/aai-2026-rocm-ai-software "새 창에서 열림")를 도입한다. 이는 개발자가 AMD 플랫폼 전반에서 GPU 소프트웨어를 더 빠르게 구축, 최적화 및 배포할 수 있도록 돕는 AI 기반 개발 플랫폼이다. ROCm.ai는 Claude, Codex, Cursor와 같은 인기 있는 코딩 에이전트가 AMD 플랫폼과 ROCm을 기본적으로 이해할 수 있게 함으로써 AI 지원 GPU 프로그래밍을 개발자에게 제공한다.
ROCm.ai는 AMD Instinct MI455X GPU의 소프트웨어 지원을 가속화하는 동시에 성능을 최적화하고 있다. PyTorch, Hugging Face, vLLM, SGLang을 포함한 선도적인 오픈소스 프레임워크는 이미 MI455X에서 지원되며 좋은 결과를 보이고 있다.
**차세대 AI 인프라 가속화**
AMD는 2030년까지 CPU, GPU, 네트워킹 및 랙스케일 혁신의 연간 케이던스를 확장하고 있다. 회사는 로드맵에 대한 새로운 세부 정보를 공유했으며, 여기에는 다음이 포함된다:
* “Zen 7” 아키텍처 기반 차세대 EPYC 서버 CPU는 2028년에 출시된다. “Florence”, “Ferrara”, “Fidenza” CPU는 밀도, 성능, 시스템 달러당 성능 및 와트당 성능에서 AMD의 리더십을 확장할 것으로 예상된다.
* “Zen 8” 아키텍처 기반 “Ravenna” CPU는 2030년에 출시되어 AMD 서버 CPU 리더십을 지속하도록 설계되었다.
* 차세대 AMD Instinct MI500 시리즈 GPU는 2027년에 출시되며, 리더십 성능을 위한 차세대 컴퓨트, 메모리 및 인터커넥트 기술을 갖춘다.
* AMD Instinct MI600 시리즈 GPU는 2028년에 출시된다.
* 차세대 AMD Helios 500 랙스케일 솔루션은 AMD Instinct MI500 시리즈 GPU와 AMD EPYC “Verano” CPU, 차세대 Pensando “Como” 및 “Monza” 네트워킹으로 구동된다. AMD Helios 600 랙스케일 솔루션은 AMD Instinct MI600 시리즈 GPU, EPYC “Ferrara” CPU 및 Pensando “Palma” 및 “Levanzo” 네트워킹으로 구동되며 뒤따를 것이다.
**엔터프라이즈 전반에 AI 확장**
선도적인 엔터프라이즈는 클라우드, 하이브리드 및 온프레미스 데이터센터에서 AI 지원 PC 플릿에 이르기까지 AMD 인프라에서 실행된다. AMD 기술은 고객이 빠르게 확장하고 엔터프라이즈 전환을 가속화하도록 돕고 있다.
Advancing AI에서 AT&T는 클라우드, 온프레미스 및 에어갭 환경 전반에 걸쳐 AMD 기술을 사용하여 유연한 엔터프라이즈 AI를 배포하는 방법을 설명했다. AT&T는 또한 AMD Instinct GPU와 ROCm 소프트웨어를 사용하여 통신용으로 특별히 훈련된 오픈소스 모델인 OTel 2.0 모델을 구동하고 있다.
AMD Ryzen™ AI Halo 개발자 플랫폼을 통해 AMD는 로컬 AI 개발을 접근 가능하게 만드는 성능, 효율성 및 단순성을 제공한다. Ryzen™ AI Max PRO 400 시리즈 프로세서로 구동되는 더 많은 AMD Ryzen AI Halo 플랫폼이 올해 말 AMD 및 OEM 파트너로부터 제공될 예정이다.
Cisco와 AMD는 AMD Ryzen AI Halo 시스템을 포함한 AMD 고성능 추론 엔진을 Cisco 네트워킹, 관찰 가능성 및 보안 기능과 결합하여 기업이 하이브리드 및 로컬 에이전틱 AI를 대규모로 배포, 관리 및 거버넌스할 수 있도록 협력하고 있다.
**피지컬 AI의 차세대 프론티어 발전**
AI가 클라우드, 엔터프라이즈 및 로컬 시스템 전반으로 확장됨에 따라 다음 프론티어는 물리적 세계에서 인식, 추론 및 행동하는 기계에 지능을 가져오는 것이다. AMD FPGA 및 적응형 SoC를 통한 로보틱스의 오랜 유산을 바탕으로 AMD는 로봇 몸체에서 로봇 두뇌로 회사의 로보틱스 역량을 확장하는 AMD Kria™ AI 솔루션을 도입했다. AMD는 AI 인식, 추론 및 에이전틱 의사 결정과 제어를 단일 플랫폼에서 함께 제공하여 까다로운 실제 로봇 시스템에 필요한 성능을 제공하는 독보적인 위치에 있다.
포트폴리오에는 새로운 [AMD Kria AI 시스템 온 모듈](https://www.amd.com/en/products/system-on-modules/kria/ai.html "새 창에서 열림")(SOM)과 새로운 [AMD Ryzen AI Embedded X100 시리즈 프로세서](https://newsroom.amd.com/news/aai-2026-ryzen-ai-embedded-x100 "새 창에서 열림")로 구동되는 것, 그리고 CPU, GPU, NPU 및 FPGA 컴퓨트를 결합한 자율 로보틱스용 최초의 개방형 턴키 통합 플랫폼인 [AMD Kria AI 로보틱스 개발자 플랫폼](https://newsroom.amd.com/news/aai-2026-kria-robotics-dev-platform "새 창에서 열림")이 포함된다. 확장된 개방형 소프트웨어 생태계와 함께 AMD Kria AI 솔루션은 벤더 종속성을 제거하고 개발자와 고객이 차세대 피지컬 AI 시스템의 프로토타입에서 생산까지 가는 경로를 가속화하도록 돕는다.
**지원 리소스**
* [Advancing AI 2026](https://newsroom.amd.com/press-kits/advancing-ai-2026-all-news "새 창에서 열림")에서 AMD 팔로우 (프레스 키트)
* [AAI 기조연설 다시보기](https://newsroom.amd.com/news/aai-2026-keynote-livestre
[원문 후반 발췌]
logy or other intellectual property; unfavorable currency exchange rate fluctuations; ability of third party manufacturers to manufacture AMD's products on a timely basis in sufficient quantities and using competitive technologies; availability of essential equipment, materials, components (such as memory supply), substrates or manufacturing processes; ability to achieve expected manufacturing yields for AMD’s products; AMD's ability to generate revenue from its semi-custom SoC products; potential security vulnerabilities; potential security incidents including IT outages, data loss, data breaches and cyberattacks; uncertainties involving the ordering and shipment of AMD’s products; AMD’s reliance on third-party intellectual property to design and introduce new products; AMD's reliance on third-party companies for design, manufacture and supply of motherboards, software, memory and other computer platform components; AMD's reliance on Microsoft and other software vendors' support to design and develop software to run on AMD’s products; AMD’s reliance on third-party distributors and add-in-board partners; impact of modification or interruption of AMD’s internal business processes and information systems; compatibility of AMD’s products with some or all industry-standard software and har
[원문 후반 발췌]
general-purpose enterprise, cloud, telecom, SMB, and HPC systems to emerging AI environments including sandboxed agentic AI deployments and GPU head node servers. AMD EPYC 6th Generation platforms extend this breadth by uniquely combining high core and thread density of up to 512 threads, advanced memory bandwidth of up to 16 channels of 12.8 GT/s MRDIMM support, next-generation PCIe® Gen 6 connectivity, and select SKUs with boost frequencies up to 5 GHz.
3 EPYC-025D: As of July 2026, 6th Gen EPYC 9996 has 256 cores and 512 threads with SMT enabled which is higher than any other publicly disclosed 1P CPU
4 9xx6-012: Based on estimated performance data for Nvidia, Intel®, and AMD EPYC™ Server Processors for Agentic AI, the AMD EPYC9996 provides the most agents per rack at a 100Kw power envelope per rack.
Compared to the cores per rack of Nvidia Vera (88c) powered server racks:
- The AMD EPYC 9996 (256C) provides 2.08x the cores (and threads with SMT) per rack
- The AMD EPYC 9965 (192C) provides 1.86x the cores(and threads with SMT) per rack
- The Intel Xeon 6980P (128C) provides 1.24x the cores(and threads with SMT) per rack
Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf
Agent counts are estimates derived from available
[원문 후반 발췌]
Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
6 9xx6-014: Comparison based on published Top-of-stack core counts and CPU W, Default CPU Power for 6th Gen EPYC, and TDPs for 5th Gen EPYC, Intel® Xeon®, and Nvidia Vera for estimated Highest Agents / CPU W across 6th Gen AMD EPYC™ 256C at 400W, AMD EPYC™ 9965 (500W TDP), Nvidia Vera (450W TDP), ARM AGI (300W TDP), Intel® Xeon® 6980P (500W TDP), and Intel Xeon 6990E+ (450W TDP) powered servers as of 7/22/2026. 2 threads per core (SMT). 1 thread per core for ARM AGI.
Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
Starting with the 6th Gen AMD EPYC™ server processor family, AMD uses Default CPU Power to describe processor power consumption, succeeding AMD's historical TDP reference. Default CPU Power reflects total power
본문이 길어 23840자 중 14960자만 처리했습니다.
브리프용 요약 초안
AMD가 Advancing AI 2026에서 Helios 랙스케일 AI 솔루션과 6세대 EPYC CPU를 발표했다. Anthropic은 최대 2GW 규모의 MI455X GPU 배치를, OpenAI는 2026년 4분기부터 Helios 온라인을 계획하고 있다. 서버 DRAM 및 HBM 수요에 긍정적일 수 있으나, 원문은 메모리 공급업체나 구체적 메모리 용량을 명시하지 않아 직접적 영향은 미확인이다.
원문 텍스트
원문 열기 ↗AMD launches broad high-performance computing and physical AI portfolios, including its first rack-scale AI solution and the world’s most powerful AI rack, AMD Helios.
**News Highlights**
* At Advancing AI 2026, AMD launches 6th Gen AMD EPYC™ CPUs, AMD Instinct™ MI400 Series GPUs, AMD Helios™ AI rackscale solutions, AMD Ryzen™ AI Embedded X100 processors and the AMD Kria™ AI SOM and Robotics Developer Platform.
* AMD Helios delivers up to 30% more inference tokens per dollar than the competition, maximizing output from every rack deployed.
* AI is accelerating demand for the full range of AMD silicon spanning data center, PCs, edge and embedded processors, driving AMD’s TAM to ~$2 trillion in 2030.
* Anthropic, OpenAI, Meta, Cerebras, AT&T and Cisco detailed how they are collaborating with AMD to advance AI infrastructure, enterprise capabilities and edge AI.
SAN FRANCISCO, July 23, 2026 (GLOBE NEWSWIRE) -- [AMD](http://www.amd.com/ "Opens in a new window") (NASDAQ: AMD) today launched its next-generation AI infrastructure and physical AI portfolio at Advancing AI 2026, led by AMD Helios rackscale solutions, now in production to be deployed by leading AI companies at gigawatt scale.
As AI expands from training to inference and agentic workloads, compute demand is accelerating rapidly. AMD delivers an open, full-stack AI platform that gives customers the flexibility to deploy the right compute for every workload.
“The next phase of AI will span frontier models, agents and physical AI, creating new opportunities to bring intelligence everywhere,” said Dr. Lisa Su, chair and CEO, AMD. “Realizing that potential will take the entire industry working together. AMD is partnering across the ecosystem to deliver leadership compute and open platforms that give customers the performance, flexibility and choice to scale AI from the data center to the edge.”
**AMD Helios: The Highest Performance Rack-Scale AI Solution**
Delivering frontier AI requires a fully integrated rack architecture, with every part of the stack pushing the boundaries of performance. [AMD Helios rackscale solutions](https://newsroom.amd.com/news/aai-2026-helios-update "Opens in a new window") are built for this, with co-optimized silicon spanning 72 high-performance AMD Instinct™ MI455X GPUs and 18 powerful 6th Gen AMD EPYC™ “Venice” CPUs, connected by AMD Pensando™ front-end, scale-up and scale-out networking, and accelerated by AMD ROCm™ open software. AMD Helios combines leadership compute performance, memory capacity and networking bandwidth to deliver up to 30% more tokens per dollar than the leading competitive solution 1.
Leading AI labs and cloud providers are choosing AMD Helios for its open, full-stack performance. They include OpenAI, Anthropic, Meta, Microsoft, Oracle, HUMAIN, Tensorwave, Vultr, Cirrascale and others. Systems will be available from leading OEMs, including Bull, HPE, Lenovo and Supermicro, as well as infrastructure partners Sanmina and Wiwynn.
At Advancing AI, AMD partners detailed how they deploy AMD AI infrastructure at scale for frontier training and inference:
* Anthropic and AMD further outlined Wednesday’s strategic partnership announcement to deploy up to 2 gigawatts of AMD Instinct MI455X GPUs in AMD Helios rackscale solutions. The companies are launching a multiyear engineering collaboration to use Claude to accelerate AMD software development. Specifically, the teams will use Claude to optimize workloads for AMD Instinct GPUs and accelerate ROCm software development. AMD will also broadly adopt Claude across its engineering and product development teams.
* OpenAI and AMD are partnering to optimize the full AI stack, from silicon to software. Leveraging OpenAI’s Triton framework with AMD ROCm software, the companies are optimizing GPT-class workloads on AMD Instinct MI455X GPUs and AMD Helios racks. OpenAI expects to bring Helios online beginning in the fourth quarter of 2026, with deployments accelerating throughout 2027.
* Meta and AMD are co-designing for gigawatt-scale deployments, optimizing AMD’s full AI compute stack for Meta workloads. Meta is now validating 6th Gen EPYC CPU platforms in its labs and has begun testing and validating workloads on AMD Helios racks as they prepare to deploy at scale.
* Cerebras and AMD are [collaborating](https://newsroom.amd.com/news/aai-2026-cerebras-inference "Opens in a new window") to deliver a combined solution of Cerebras ultra-low-latency AI compute and AMD Helios high-throughput rack-scale infrastructure to help improve inference efficiency, scalability and economics for ultra-low-latency inference serving.
**Delivering the Highest Performance Data Center CPUs and GPUs**
6th Gen EPYC processors deliver the broadest server CPU portfolio for agentic AI,2 spanning cloud, enterprise, general-purpose and high-performance computing (HPC) workloads. With leadership per-core performance and the highest thread density 3, they enable the most agents per watt, per dollar and per rack.4,5,6 For AI host nodes, [6th Gen EPYC CPUs](https://newsroom.amd.com/news/aai-2026-6th-gen-epyc "Opens in a new window") deliver the speed and memory bandwidth to keep accelerators fully fed. And for general-purpose servers, they bring leadership performance and energy efficiency to run business critical applications and AI support tasks.
With [AMD Instinct™ MI400 Series GPUs](https://newsroom.amd.com/news/aai-2026-mi400-instinct-update "Opens in a new window"), AMD delivers powerful performance for cloud, enterprise and HPC workloads. AMD Instinct MI455X GPUs deliver 34x higher token throughput compared to MI355X GPUs 7. For high-precision workloads, the AMD Instinct™ MI430X accelerator is the most advanced for HPC and sovereign AI with up to 288 TFLOPS of hardware-based FP64 performance for scientific computing. Instinct MI430X accelerators are powering the next wave of exascale-class supercomputers across the U.S. and Europe.
AMD also launched the Instinct MI350P GPU, bringing seamless AI acceleration to existing infrastructure with leadership token economics. MI350P GPUs deliver up to 4.2x more tokens per second per dollar than the competition 8.
**Advancing the Open Software Ecosystem**
For developers, AMD ROCm is the open software platform with the performance, flexibility and ecosystem support needed to build and deploy AI on AMD hardware. Building on that foundation, AMD is introducing [ROCm.ai](https://newsroom.amd.com/news/aai-2026-rocm-ai-software "Opens in a new window"), an AI-driven development platform that helps developers build, optimize and deploy GPU software faster across AMD platforms. ROCm.ai brings AI-assisted GPU programming to developers by enabling popular coding agents such as Claude, Codex and Cursor to understand AMD platforms and ROCm natively.
ROCm.ai is accelerating software enablement for AMD Instinct MI455X GPUs while optimizing performance. Leading open-source frameworks including PyTorch, Hugging Face, vLLM and SGLang are already enabled on MI455X and seeing great results.
**Accelerating Next-Generation AI Infrastructure**
AMD is extending its annual cadence of CPU, GPU, networking and rack-scale innovation through 2030. The company shared new details on its roadmaps, including:
* Next-generation EPYC server CPUs based on the “Zen 7” architecture are coming in 2028. The “Florence,” “Ferrara” and “Fidenza” CPUs are expected to extend AMD’s leadership in density, performance, performance-per-system dollar and performance-per-watt.
* “Ravenna” CPUs based on the “Zen 8” architecture are coming in 2030, designed to continue AMD server CPU leadership.
* Next-generation AMD Instinct MI500 Series GPUs are coming in 2027, with next-generation compute, memory and interconnect technologies for leadership performance.
* AMD Instinct MI600 Series GPUs are coming in 2028.
* The next-generation AMD Helios 500 rackscale solution will be powered by AMD Instinct MI500 Series GPUs and AMD EPYC “Verano” CPUs, with next-gen Pensando “Como” and “Monza” networking. The AMD Helios 600 rackscale solution will follow, powered by AMD Instinct MI600 Series GPUs, EPYC “Ferrara” CPUs and Pensando “Palma” and “Levanzo” networking.
**Scaling AI Across Enterprise**
Leading enterprises run on AMD infrastructure, from cloud, hybrid and on-prem data centers to AI-enabled PC fleets. AMD technologies are helping customers scale quickly and accelerate enterprise transformation.
At Advancing AI, AT&T illustrated how it is deploying flexible enterprise AI using AMD technology across cloud, on-premises and air-gapped environments. AT&T is also using AMD Instinct GPUs and ROCm software to power its OTel 2.0 model, an open-source model trained specifically for telecoms.
With the AMD Ryzen™ AI Halo developer platform, AMD delivers performance, efficiency and simplicity that makes local AI development accessible. More AMD Ryzen AI Halo platforms, powered by Ryzen™ AI Max PRO 400 Series processors, will be available later this year from AMD and OEM partners.
Cisco and AMD are collaborating to combine AMD high-performance inference engines, including AMD Ryzen AI Halo systems, with Cisco networking, observability and security capabilities, so enterprises can deploy, govern and manage hybrid and local agentic AI at scale.
**Advancing the Next Frontier of Physical AI**
As AI expands across cloud, enterprise and local systems, the next frontier is bringing intelligence into machines that perceive, reason and act in the physical world. Building on a long legacy in robotics with AMD FPGAs and adaptive SoCs, AMD introduced AMD Kria™ AI solutions, extending the company’s robotics capabilities from the robot body to the robot brain. AMD uniquely brings AI perception, reasoning and agentic decision-making and control together on a single platform to deliver the performance required for demanding real-world robotic systems.
The portfolio includes new [AMD Kria AI system-on-modules](https://www.amd.com/en/products/system-on-modules/kria/ai.html "Opens in a new window") (SOMs), powered by the new [AMD Ryzen AI Embedded X100 Series processors](https://newsroom.amd.com/news/aai-2026-ryzen-ai-embedded-x100 "Opens in a new window"), and the [AMD Kria AI Robotics Developer Platform](https://newsroom.amd.com/news/aai-2026-kria-robotics-dev-platform "Opens in a new window"), the first open, turnkey integrated platform for autonomous robotics combining CPU, GPU, NPU and FPGA compute. Together with an expanded open software ecosystem, AMD Kria AI solutions remove vendor lock-in and help developers and customers accelerate the path from prototype to production for next-generation physical AI systems.
**Supporting Resources**
* Follow AMD at [Advancing AI 2026](https://newsroom.amd.com/press-kits/advancing-ai-2026-all-news "Opens in a new window") (Press Kit)
* Watch the [AAI keynote replay](https://newsroom.amd.com/news/aai-2026-keynote-livestream-replay "Opens in a new window")
**About AMD**
AMD (NASDAQ: AMD) drives innovation in high-performance and AI computing to solve the world’s most important challenges. Today, AMD technology powers billions of experiences across cloud and AI infrastructure, embedded systems, AI PCs and gaming. With a broad portfolio of AI-optimized CPUs, GPUs, networking and software, AMD delivers full-stack AI solutions that provide the performance and scalability needed for a new era of intelligent computing. Learn more at[www.amd.com](http://www.amd.com/ "Opens in a new window").
**CAUTIONARY STATEMENT**
This press release contains forward-looking statements concerning Advanced Micro Devices, Inc. (AMD) such as AI accelerating demand for the full range of AMD silicon, driving AMD’s total addressable market to ~$2 trillion in 2030; compute demand accelerating; the next phase of AI; the features, functionality, performance, availability, timing and expected benefits of AMD products, including, but not limited to, 6th Gen AMD EPYC™ CPUs, AMD Instinct™ MI400 Series GPUs, AMD Helios™ AI rackscale solutions, AMD Ryzen™ AI Embedded X100 processors and the AMD Kria™ AI SOM and Robotics Developer Platform; expected plans, benefits, scalability and deployments with AMD partners including Anthropic, OpenAI, Meta and Cerebras; AMD’s AI infrastructure product roadmaps through 2030; and AMD’s collaboration with AT&T and Cisco, 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 the forward-looking statements in this press release are based on current beliefs, assumptions and expectations, speak only as of the date of this press release 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 are 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. Material factors that could cause actual results to differ materially from current expectations include, without limitation, the following: impact of government actions and regulations such as export regulations, import tariffs, trade protection measures, and licensing requirements; competitive markets in which AMD’s products are sold; the cyclical nature of the semiconductor industry; market conditions of the industries in which AMD products are sold; AMD’s ability to introduce products on a timely basis with expected features and performance levels; loss of a significant customer; economic and market uncertainty; quarterly and seasonal sales patterns; AMD's ability to adequately protect its technology or other intellectual property; unfavorable currency exchange rate fluctuations; ability of third party manufacturers to manufacture AMD's products on a timely basis in sufficient quantities and using competitive technologies; availability of essential equipment, materials, components (such as memory supply), substrates or manufacturing processes; ability to achieve expected manufacturing yields for AMD’s products; AMD's ability to generate revenue from its semi-custom SoC products; potential security vulnerabilities; potential security incidents including IT outages, data loss, data breaches and cyberattacks; uncertainties involving the ordering and shipment of AMD’s products; AMD’s reliance on third-party intellectual property to design and introduce new products; AMD's reliance on third-party companies for design, manufacture and supply of motherboards, software, memory and other computer platform components; AMD's reliance on Microsoft and other software vendors' support to design and develop software to run on AMD’s products; AMD’s reliance on third-party distributors and add-in-board partners; impact of modification or interruption of AMD’s internal business processes and information systems; compatibility of AMD’s products with some or all industry-standard software and hardware; costs related to defective products; failure to maintain an efficient supply chain as customer demand changes; AMD's ability to rely on third party supply-chain logistics functions; AMD’s ability to effectively control sales of its products on the gray market; impact of climate change on AMD’s business; AMD’s ability to realize its deferred tax assets; potential tax liabilities; current and future claims and litigation; impact of environmental laws, conflict minerals related provisions and other laws or regulations; evolving expectations from governments, investors, customers and other stakeholders regarding corporate responsibility matters; issues related to the responsible use of AI; restrictions imposed by agreements governing AMD’s notes, the guarantees of Xilinx’s notes and the revolving credit agreement; AMD’s ability to satisfy financial obligations under guarantees, leases and other commercial commitments; impact of acquisitions, joint ventures and/or investments on AMD’s business and AMD’s ability to integrate acquired businesses; impact of any impairment of the combined company’s assets; political, legal and economic risks and natural disasters; future impairments of technology license purchases; AMD’s ability to attract and retain key employees; and AMD’s stock price volatility. 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.
_________________________
1 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
2 EPYC-068 - The AMD EPYC server CPU portfolio spans the industry’s broadest ranges of data center deployments, from general-purpose enterprise, cloud, telecom, SMB, and HPC systems to emerging AI environments including sandboxed agentic AI deployments and GPU head node servers. AMD EPYC 6th Generation platforms extend this breadth by uniquely combining high core and thread density of up to 512 threads, advanced memory bandwidth of up to 16 channels of 12.8 GT/s MRDIMM support, next-generation PCIe® Gen 6 connectivity, and select SKUs with boost frequencies up to 5 GHz.
3 EPYC-025D: As of July 2026, 6th Gen EPYC 9996 has 256 cores and 512 threads with SMT enabled which is higher than any other publicly disclosed 1P CPU
4 9xx6-012: Based on estimated performance data for Nvidia, Intel®, and AMD EPYC™ Server Processors for Agentic AI, the AMD EPYC9996 provides the most agents per rack at a 100Kw power envelope per rack.
Compared to the cores per rack of Nvidia Vera (88c) powered server racks:
- The AMD EPYC 9996 (256C) provides 2.08x the cores (and threads with SMT) per rack
- The AMD EPYC 9965 (192C) provides 1.86x the cores(and threads with SMT) per rack
- The Intel Xeon 6980P (128C) provides 1.24x the cores(and threads with SMT) per rack
Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf
Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
5 9xx6-013: Comparison based on published Top-of-stack core counts and 1Ku pricing for estimated highest Agents / CPU $ and threads / CPU $ AMD EPYC™ 9006 (512 threads), AMD EPYC™ 9005 (384 threads), and Intel® Xeon® 6 (256 threads) SKUs as of 7/22/2026. Threads derived as 2 threads per core (SMT). Intel and Xeon are trademarks of Intel Corporation or its subsidiaries. Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
6 9xx6-014: Comparison based on published Top-of-stack core counts and CPU W, Default CPU Power for 6th Gen EPYC, and TDPs for 5th Gen EPYC, Intel® Xeon®, and Nvidia Vera for estimated Highest Agents / CPU W across 6th Gen AMD EPYC™ 256C at 400W, AMD EPYC™ 9965 (500W TDP), Nvidia Vera (450W TDP), ARM AGI (300W TDP), Intel® Xeon® 6980P (500W TDP), and Intel Xeon 6990E+ (450W TDP) powered servers as of 7/22/2026. 2 threads per core (SMT). 1 thread per core for ARM AGI.
Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
Starting with the 6th Gen AMD EPYC™ server processor family, AMD uses Default CPU Power to describe processor power consumption, succeeding AMD's historical TDP reference. Default CPU Power reflects total power consumed across the processor's compute and I/O dies for the stated performance target. Default CPU Power and TDP may both serve as processor power references for product comparison, platform planning, and performance-per-watt analysis. Intel Xeon TDP from [ark.intel.com](https://www.intel.com/content/www/us/en/ark.html "Opens in a new window"). Nvidia Vera TDP from [https://developer.nvidia.com/blog/nvidia-vera-cpu-sets-a-new-standard-for-agentic-workloads-in-ai-factories](https://developer.nvidia.com/blog/nvidia-vera-cpu-sets-a-new-standard-for-agentic-workloads-in-ai-factories/ "Opens in a new window")/. ARM AGI Specifications from [https://www.arm.com/products/cloud-datacenter/arm-agi-cpu#Specifications](https://www.arm.com/products/cloud-datacenter/arm-agi-cpu "Opens in a new window")
7 MI400-020: Based on measurements and calculations by AMD Performance Labs in July 2026, for the AMD Instinct™ MI455X GPU to determine measured token throughput at high, medium and low interactivity points run on Deepseek V4 Flash with FP4 serving compared to AMD Instinct™ MI355X GPU. System manufacturers may vary configurations, yielding different results.
8 MI350P-007: Based on AMD internal testing (July 2026), on a (1x) AMD Instinct MI350P GPU vs (1x) NVIDIA H200 NVL GPU on the Llama 3.3 70B Instruct (FP8) online serving output-throughput per dollar (tok/s/USD) comparison at ISL/OSL 1024/1024 across concurrency levels 1, 4, 8, 16, 32, 64, 128, 256, 512; median of 3 runs per point. MI350P based server internal AMD estimated pricing as $327,238.40 USD. RTX_PRO_6000 based server public list price reported on OEM website as $265,928.24 USD as of 7/16/2026. Stated results are the peak per-concurrency ratios: MI350P served via AIMS silogenai/aim-instinct-meta-llama-llama-3-3-70b-instruct:0.12.0-rc6; H200 NVL via NVIDIA NIM nvcr.io/nim/meta/llama-3.3-70b-instruct:2.0.6; RTX PRO 6000 via NVIDIA NIM nvcr.io/nim/meta/llama-3.3-70b-instruct:2.0.6. Configuration: 8x AMD Instinct MI350P PCIe Card (CDNA4, gfx950, 128 CUs, 144 GB HBM3E, SPX compute / NPS1), vBIOS 113-350P-01-1K1-000A, GPU driver 6.19.13-2353916.24.04, ROCm 7.14.0 (AMD-SMI 26.5.0); host 2P AMD EPYC 9455 (48-core), Dell PowerEdge XE7745, BIOS 1.7.6, microcode 0xb002162, SMT Enabled, Ubuntu 24.04.4 LTS, Linux 6.8.0-124-generic || NVIDIA RTX PRO 6000: 8x NVIDIA RTX PRO 6000 Blackwell Server Edition, vBIOS 98.02.8D.00.01, GPU driver 595.45.04, CUDA 13.2; host 2P AMD EPYC 9455 (48-core), Dell PowerEdge XE7745, BIOS 1.6.4, microcode 0xb00215a, SMT Enabled, Ubuntu 24.04.4 LTS, Linux 6.8.0-124-generic. Sever manufacturers may vary configurations, yielding different results. Results may vary due to factors including system configurations, software versions and BIOS settings.
**Contact:**
**Brandi Martina**
AMD Communications
(512) 705-1720
Brandi.martina@amd.com
**Liz Stine**
AMD Investor Relations
(720) 652-3965
liz.stine@amd.com
[](https://www.globenewswire.com/NewsRoom/AttachmentNg/9782a337-07ac-424b-9c5a-f9cde47d8d55)
Source: Advanced Micro Devices, Inc.
Released July 23, 2026
**News Highlights**
* At Advancing AI 2026, AMD launches 6th Gen AMD EPYC™ CPUs, AMD Instinct™ MI400 Series GPUs, AMD Helios™ AI rackscale solutions, AMD Ryzen™ AI Embedded X100 processors and the AMD Kria™ AI SOM and Robotics Developer Platform.
* AMD Helios delivers up to 30% more inference tokens per dollar than the competition, maximizing output from every rack deployed.
* AI is accelerating demand for the full range of AMD silicon spanning data center, PCs, edge and embedded processors, driving AMD’s TAM to ~$2 trillion in 2030.
* Anthropic, OpenAI, Meta, Cerebras, AT&T and Cisco detailed how they are collaborating with AMD to advance AI infrastructure, enterprise capabilities and edge AI.
SAN FRANCISCO, July 23, 2026 (GLOBE NEWSWIRE) -- [AMD](http://www.amd.com/ "Opens in a new window") (NASDAQ: AMD) today launched its next-generation AI infrastructure and physical AI portfolio at Advancing AI 2026, led by AMD Helios rackscale solutions, now in production to be deployed by leading AI companies at gigawatt scale.
As AI expands from training to inference and agentic workloads, compute demand is accelerating rapidly. AMD delivers an open, full-stack AI platform that gives customers the flexibility to deploy the right compute for every workload.
“The next phase of AI will span frontier models, agents and physical AI, creating new opportunities to bring intelligence everywhere,” said Dr. Lisa Su, chair and CEO, AMD. “Realizing that potential will take the entire industry working together. AMD is partnering across the ecosystem to deliver leadership compute and open platforms that give customers the performance, flexibility and choice to scale AI from the data center to the edge.”
**AMD Helios: The Highest Performance Rack-Scale AI Solution**
Delivering frontier AI requires a fully integrated rack architecture, with every part of the stack pushing the boundaries of performance. [AMD Helios rackscale solutions](https://newsroom.amd.com/news/aai-2026-helios-update "Opens in a new window") are built for this, with co-optimized silicon spanning 72 high-performance AMD Instinct™ MI455X GPUs and 18 powerful 6th Gen AMD EPYC™ “Venice” CPUs, connected by AMD Pensando™ front-end, scale-up and scale-out networking, and accelerated by AMD ROCm™ open software. AMD Helios combines leadership compute performance, memory capacity and networking bandwidth to deliver up to 30% more tokens per dollar than the leading competitive solution 1.
Leading AI labs and cloud providers are choosing AMD Helios for its open, full-stack performance. They include OpenAI, Anthropic, Meta, Microsoft, Oracle, HUMAIN, Tensorwave, Vultr, Cirrascale and others. Systems will be available from leading OEMs, including Bull, HPE, Lenovo and Supermicro, as well as infrastructure partners Sanmina and Wiwynn.
At Advancing AI, AMD partners detailed how they deploy AMD AI infrastructure at scale for frontier training and inference:
* Anthropic and AMD further outlined Wednesday’s strategic partnership announcement to deploy up to 2 gigawatts of AMD Instinct MI455X GPUs in AMD Helios rackscale solutions. The companies are launching a multiyear engineering collaboration to use Claude to accelerate AMD software development. Specifically, the teams will use Claude to optimize workloads for AMD Instinct GPUs and accelerate ROCm software development. AMD will also broadly adopt Claude across its engineering and product development teams.
* OpenAI and AMD are partnering to optimize the full AI stack, from silicon to software. Leveraging OpenAI’s Triton framework with AMD ROCm software, the companies are optimizing GPT-class workloads on AMD Instinct MI455X GPUs and AMD Helios racks. OpenAI expects to bring Helios online beginning in the fourth quarter of 2026, with deployments accelerating throughout 2027.
* Meta and AMD are co-designing for gigawatt-scale deployments, optimizing AMD’s full AI compute stack for Meta workloads. Meta is now validating 6th Gen EPYC CPU platforms in its labs and has begun testing and validating workloads on AMD Helios racks as they prepare to deploy at scale.
* Cerebras and AMD are [collaborating](https://newsroom.amd.com/news/aai-2026-cerebras-inference "Opens in a new window") to deliver a combined solution of Cerebras ultra-low-latency AI compute and AMD Helios high-throughput rack-scale infrastructure to help improve inference efficiency, scalability and economics for ultra-low-latency inference serving.
**Delivering the Highest Performance Data Center CPUs and GPUs**
6th Gen EPYC processors deliver the broadest server CPU portfolio for agentic AI,2 spanning cloud, enterprise, general-purpose and high-performance computing (HPC) workloads. With leadership per-core performance and the highest thread density 3, they enable the most agents per watt, per dollar and per rack.4,5,6 For AI host nodes, [6th Gen EPYC CPUs](https://newsroom.amd.com/news/aai-2026-6th-gen-epyc "Opens in a new window") deliver the speed and memory bandwidth to keep accelerators fully fed. And for general-purpose servers, they bring leadership performance and energy efficiency to run business critical applications and AI support tasks.
With [AMD Instinct™ MI400 Series GPUs](https://newsroom.amd.com/news/aai-2026-mi400-instinct-update "Opens in a new window"), AMD delivers powerful performance for cloud, enterprise and HPC workloads. AMD Instinct MI455X GPUs deliver 34x higher token throughput compared to MI355X GPUs 7. For high-precision workloads, the AMD Instinct™ MI430X accelerator is the most advanced for HPC and sovereign AI with up to 288 TFLOPS of hardware-based FP64 performance for scientific computing. Instinct MI430X accelerators are powering the next wave of exascale-class supercomputers across the U.S. and Europe.
AMD also launched the Instinct MI350P GPU, bringing seamless AI acceleration to existing infrastructure with leadership token economics. MI350P GPUs deliver up to 4.2x more tokens per second per dollar than the competition 8.
**Advancing the Open Software Ecosystem**
For developers, AMD ROCm is the open software platform with the performance, flexibility and ecosystem support needed to build and deploy AI on AMD hardware. Building on that foundation, AMD is introducing [ROCm.ai](https://newsroom.amd.com/news/aai-2026-rocm-ai-software "Opens in a new window"), an AI-driven development platform that helps developers build, optimize and deploy GPU software faster across AMD platforms. ROCm.ai brings AI-assisted GPU programming to developers by enabling popular coding agents such as Claude, Codex and Cursor to understand AMD platforms and ROCm natively.
ROCm.ai is accelerating software enablement for AMD Instinct MI455X GPUs while optimizing performance. Leading open-source frameworks including PyTorch, Hugging Face, vLLM and SGLang are already enabled on MI455X and seeing great results.
**Accelerating Next-Generation AI Infrastructure**
AMD is extending its annual cadence of CPU, GPU, networking and rack-scale innovation through 2030. The company shared new details on its roadmaps, including:
* Next-generation EPYC server CPUs based on the “Zen 7” architecture are coming in 2028. The “Florence,” “Ferrara” and “Fidenza” CPUs are expected to extend AMD’s leadership in density, performance, performance-per-system dollar and performance-per-watt.
* “Ravenna” CPUs based on the “Zen 8” architecture are coming in 2030, designed to continue AMD server CPU leadership.
* Next-generation AMD Instinct MI500 Series GPUs are coming in 2027, with next-generation compute, memory and interconnect technologies for leadership performance.
* AMD Instinct MI600 Series GPUs are coming in 2028.
* The next-generation AMD Helios 500 rackscale solution will be powered by AMD Instinct MI500 Series GPUs and AMD EPYC “Verano” CPUs, with next-gen Pensando “Como” and “Monza” networking. The AMD Helios 600 rackscale solution will follow, powered by AMD Instinct MI600 Series GPUs, EPYC “Ferrara” CPUs and Pensando “Palma” and “Levanzo” networking.
**Scaling AI Across Enterprise**
Leading enterprises run on AMD infrastructure, from cloud, hybrid and on-prem data centers to AI-enabled PC fleets. AMD technologies are helping customers scale quickly and accelerate enterprise transformation.
At Advancing AI, AT&T illustrated how it is deploying flexible enterprise AI using AMD technology across cloud, on-premises and air-gapped environments. AT&T is also using AMD Instinct GPUs and ROCm software to power its OTel 2.0 model, an open-source model trained specifically for telecoms.
With the AMD Ryzen™ AI Halo developer platform, AMD delivers performance, efficiency and simplicity that makes local AI development accessible. More AMD Ryzen AI Halo platforms, powered by Ryzen™ AI Max PRO 400 Series processors, will be available later this year from AMD and OEM partners.
Cisco and AMD are collaborating to combine AMD high-performance inference engines, including AMD Ryzen AI Halo systems, with Cisco networking, observability and security capabilities, so enterprises can deploy, govern and manage hybrid and local agentic AI at scale.
**Advancing the Next Frontier of Physical AI**
As AI expands across cloud, enterprise and local systems, the next frontier is bringing intelligence into machines that perceive, reason and act in the physical world. Building on a long legacy in robotics with AMD FPGAs and adaptive SoCs, AMD introduced AMD Kria™ AI solutions, extending the company’s robotics capabilities from the robot body to the robot brain. AMD uniquely brings AI perception, reasoning and agentic decision-making and control together on a single platform to deliver the performance required for demanding real-world robotic systems.
The portfolio includes new [AMD Kria AI system-on-modules](https://www.amd.com/en/products/system-on-modules/kria/ai.html "Opens in a new window") (SOMs), powered by the new [AMD Ryzen AI Embedded X100 Series processors](https://newsroom.amd.com/news/aai-2026-ryzen-ai-embedded-x100 "Opens in a new window"), and the [AMD Kria AI Robotics Developer Platform](https://newsroom.amd.com/news/aai-2026-kria-robotics-dev-platform "Opens in a new window"), the first open, turnkey integrated platform for autonomous robotics combining CPU, GPU, NPU and FPGA compute. Together with an expanded open software ecosystem, AMD Kria AI solutions remove vendor lock-in and help developers and customers accelerate the path from prototype to production for next-generation physical AI systems.
**Supporting Resources**
* Follow AMD at [Advancing AI 2026](https://newsroom.amd.com/press-kits/advancing-ai-2026-all-news "Opens in a new window") (Press Kit)
* Watch the [AAI keynote replay](https://newsroom.amd.com/news/aai-2026-keynote-livestream-replay "Opens in a new window")
**About AMD**
AMD (NASDAQ: AMD) drives innovation in high-performance and AI computing to solve the world’s most important challenges. Today, AMD technology powers billions of experiences across cloud and AI infrastructure, embedded systems, AI PCs and gaming. With a broad portfolio of AI-optimized CPUs, GPUs, networking and software, AMD delivers full-stack AI solutions that provide the performance and scalability needed for a new era of intelligent computing. Learn more at[www.amd.com](http://www.amd.com/ "Opens in a new window").
**CAUTIONARY STATEMENT**
This press release contains forward-looking statements concerning Advanced Micro Devices, Inc. (AMD) such as AI accelerating demand for the full range of AMD silicon, driving AMD’s total addressable market to ~$2 trillion in 2030; compute demand accelerating; the next phase of AI; the features, functionality, performance, availability, timing and expected benefits of AMD products, including, but not limited to, 6th Gen AMD EPYC™ CPUs, AMD Instinct™ MI400 Series GPUs, AMD Helios™ AI rackscale solutions, AMD Ryzen™ AI Embedded X100 processors and the AMD Kria™ AI SOM and Robotics Developer Platform; expected plans, benefits, scalability and deployments with AMD partners including Anthropic, OpenAI, Meta and Cerebras; AMD’s AI infrastructure product roadmaps through 2030; and AMD’s collaboration with AT&T and Cisco, 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 the forward-looking statements in this press release are based on current beliefs, assumptions and expectations, speak only as of the date of this press release 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 are 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. Material factors that could cause actual results to differ materially from current expectations include, without limitation, the following: impact of government actions and regulations such as export regulations, import tariffs, trade protection measures, and licensing requirements; competitive markets in which AMD’s products are sold; the cyclical nature of the semiconductor industry; market conditions of the industries in which AMD products are sold; AMD’s ability to introduce products on a timely basis with expected features and performance levels; loss of a significant customer; economic and market uncertainty; quarterly and seasonal sales patterns; AMD's ability to adequately protect its technology or other intellectual property; unfavorable currency exchange rate fluctuations; ability of third party manufacturers to manufacture AMD's products on a timely basis in sufficient quantities and using competitive technologies; availability of essential equipment, materials, components (such as memory supply), substrates or manufacturing processes; ability to achieve expected manufacturing yields for AMD’s products; AMD's ability to generate revenue from its semi-custom SoC products; potential security vulnerabilities; potential security incidents including IT outages, data loss, data breaches and cyberattacks; uncertainties involving the ordering and shipment of AMD’s products; AMD’s reliance on third-party intellectual property to design and introduce new products; AMD's reliance on third-party companies for design, manufacture and supply of motherboards, software, memory and other computer platform components; AMD's reliance on Microsoft and other software vendors' support to design and develop software to run on AMD’s products; AMD’s reliance on third-party distributors and add-in-board partners; impact of modification or interruption of AMD’s internal business processes and information systems; compatibility of AMD’s products with some or all industry-standard software and hardware; costs related to defective products; failure to maintain an efficient supply chain as customer demand changes; AMD's ability to rely on third party supply-chain logistics functions; AMD’s ability to effectively control sales of its products on the gray market; impact of climate change on AMD’s business; AMD’s ability to realize its deferred tax assets; potential tax liabilities; current and future claims and litigation; impact of environmental laws, conflict minerals related provisions and other laws or regulations; evolving expectations from governments, investors, customers and other stakeholders regarding corporate responsibility matters; issues related to the responsible use of AI; restrictions imposed by agreements governing AMD’s notes, the guarantees of Xilinx’s notes and the revolving credit agreement; AMD’s ability to satisfy financial obligations under guarantees, leases and other commercial commitments; impact of acquisitions, joint ventures and/or investments on AMD’s business and AMD’s ability to integrate acquired businesses; impact of any impairment of the combined company’s assets; political, legal and economic risks and natural disasters; future impairments of technology license purchases; AMD’s ability to attract and retain key employees; and AMD’s stock price volatility. 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.
_________________________
1 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
2 EPYC-068 - The AMD EPYC server CPU portfolio spans the industry’s broadest ranges of data center deployments, from general-purpose enterprise, cloud, telecom, SMB, and HPC systems to emerging AI environments including sandboxed agentic AI deployments and GPU head node servers. AMD EPYC 6th Generation platforms extend this breadth by uniquely combining high core and thread density of up to 512 threads, advanced memory bandwidth of up to 16 channels of 12.8 GT/s MRDIMM support, next-generation PCIe® Gen 6 connectivity, and select SKUs with boost frequencies up to 5 GHz.
3 EPYC-025D: As of July 2026, 6th Gen EPYC 9996 has 256 cores and 512 threads with SMT enabled which is higher than any other publicly disclosed 1P CPU
4 9xx6-012: Based on estimated performance data for Nvidia, Intel®, and AMD EPYC™ Server Processors for Agentic AI, the AMD EPYC9996 provides the most agents per rack at a 100Kw power envelope per rack.
Compared to the cores per rack of Nvidia Vera (88c) powered server racks:
- The AMD EPYC 9996 (256C) provides 2.08x the cores (and threads with SMT) per rack
- The AMD EPYC 9965 (192C) provides 1.86x the cores(and threads with SMT) per rack
- The Intel Xeon 6980P (128C) provides 1.24x the cores(and threads with SMT) per rack
Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf
Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
5 9xx6-013: Comparison based on published Top-of-stack core counts and 1Ku pricing for estimated highest Agents / CPU $ and threads / CPU $ AMD EPYC™ 9006 (512 threads), AMD EPYC™ 9005 (384 threads), and Intel® Xeon® 6 (256 threads) SKUs as of 7/22/2026. Threads derived as 2 threads per core (SMT). Intel and Xeon are trademarks of Intel Corporation or its subsidiaries. Source: https://www.amd.com/content/dam/amd/en/documents/solutions/ai/methodology-description.pdf Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
6 9xx6-014: Comparison based on published Top-of-stack core counts and CPU W, Default CPU Power for 6th Gen EPYC, and TDPs for 5th Gen EPYC, Intel® Xeon®, and Nvidia Vera for estimated Highest Agents / CPU W across 6th Gen AMD EPYC™ 256C at 400W, AMD EPYC™ 9965 (500W TDP), Nvidia Vera (450W TDP), ARM AGI (300W TDP), Intel® Xeon® 6980P (500W TDP), and Intel Xeon 6990E+ (450W TDP) powered servers as of 7/22/2026. 2 threads per core (SMT). 1 thread per core for ARM AGI.
Agent counts are estimates derived from available CPU thread resources used as a proxy under a consistent theoretical workload. Actual agent capacity and throughput will vary based on workload, model, memory, software, orchestration, and system configuration.
Starting with the 6th Gen AMD EPYC™ server processor family, AMD uses Default CPU Power to describe processor power consumption, succeeding AMD's historical TDP reference. Default CPU Power reflects total power consumed across the processor's compute and I/O dies for the stated performance target. Default CPU Power and TDP may both serve as processor power references for product comparison, platform planning, and performance-per-watt analysis. Intel Xeon TDP from [ark.intel.com](https://www.intel.com/content/www/us/en/ark.html "Opens in a new window"). Nvidia Vera TDP from [https://developer.nvidia.com/blog/nvidia-vera-cpu-sets-a-new-standard-for-agentic-workloads-in-ai-factories](https://developer.nvidia.com/blog/nvidia-vera-cpu-sets-a-new-standard-for-agentic-workloads-in-ai-factories/ "Opens in a new window")/. ARM AGI Specifications from [https://www.arm.com/products/cloud-datacenter/arm-agi-cpu#Specifications](https://www.arm.com/products/cloud-datacenter/arm-agi-cpu "Opens in a new window")
7 MI400-020: Based on measurements and calculations by AMD Performance Labs in July 2026, for the AMD Instinct™ MI455X GPU to determine measured token throughput at high, medium and low interactivity points run on Deepseek V4 Flash with FP4 serving compared to AMD Instinct™ MI355X GPU. System manufacturers may vary configurations, yielding different results.
8 MI350P-007: Based on AMD internal testing (July 2026), on a (1x) AMD Instinct MI350P GPU vs (1x) NVIDIA H200 NVL GPU on the Llama 3.3 70B Instruct (FP8) online serving output-throughput per dollar (tok/s/USD) comparison at ISL/OSL 1024/1024 across concurrency levels 1, 4, 8, 16, 32, 64, 128, 256, 512; median of 3 runs per point. MI350P based server internal AMD estimated pricing as $327,238.40 USD. RTX_PRO_6000 based server public list price reported on OEM website as $265,928.24 USD as of 7/16/2026. Stated results are the peak per-concurrency ratios: MI350P served via AIMS silogenai/aim-instinct-meta-llama-llama-3-3-70b-instruct:0.12.0-rc6; H200 NVL via NVIDIA NIM nvcr.io/nim/meta/llama-3.3-70b-instruct:2.0.6; RTX PRO 6000 via NVIDIA NIM nvcr.io/nim/meta/llama-3.3-70b-instruct:2.0.6. Configuration: 8x AMD Instinct MI350P PCIe Card (CDNA4, gfx950, 128 CUs, 144 GB HBM3E, SPX compute / NPS1), vBIOS 113-350P-01-1K1-000A, GPU driver 6.19.13-2353916.24.04, ROCm 7.14.0 (AMD-SMI 26.5.0); host 2P AMD EPYC 9455 (48-core), Dell PowerEdge XE7745, BIOS 1.7.6, microcode 0xb002162, SMT Enabled, Ubuntu 24.04.4 LTS, Linux 6.8.0-124-generic || NVIDIA RTX PRO 6000: 8x NVIDIA RTX PRO 6000 Blackwell Server Edition, vBIOS 98.02.8D.00.01, GPU driver 595.45.04, CUDA 13.2; host 2P AMD EPYC 9455 (48-core), Dell PowerEdge XE7745, BIOS 1.6.4, microcode 0xb00215a, SMT Enabled, Ubuntu 24.04.4 LTS, Linux 6.8.0-124-generic. Sever manufacturers may vary configurations, yielding different results. Results may vary due to factors including system configurations, software versions and BIOS settings.
**Contact:**
**Brandi Martina**
AMD Communications
(512) 705-1720
Brandi.martina@amd.com
**Liz Stine**
AMD Investor Relations
(720) 652-3965
liz.stine@amd.com
[](https://www.globenewswire.com/NewsRoom/AttachmentNg/9782a337-07ac-424b-9c5a-f9cde47d8d55)
Source: Advanced Micro Devices, Inc.
Released July 23, 2026