Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

Building AI infrastructure at yotta-scale requires more than raw performance; it demands an open, modular rack design that can evolve across product generations, combining leadership compute engines with high-speed networking to connect thousands of accelerators into a single, unified system. Take a look at the Helios. Our rack-scale AI... show more

AMD

1,277,938 subscribers

20,832 просмотров • 8 месяцев назад •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

$AMD is becoming a two-engine AI story with EPYC CPUs for the growing compute needed around agentic workloads and Instinct GPUs for hyperscalers that need a credible alternative to $NVDA. The CPU side still looks underappreciated as AI agents require more high-core-count server compute and 2nm Venice should help AMD keep taking share as $INTC works through roadmap and manufacturing challenges. The bridge is Helios where AMD can package EPYC CPUs, Instinct GPUs and Pensando networking into a full rack-scale AI platform backed by real demand from $META, $ORCL and OpenAI.

$AMD is becoming a two-engine AI story with EPYC CPUs for the growing compute needed around agentic workloads and Instinct GPUs for hyperscalers that need a credible alternative to $NVDA. The CPU side still looks underappreciated as AI agents require more high-core-count server compute and 2nm Venice should help AMD keep taking share as $INTC works through roadmap and manufacturing challenges. The bridge is Helios where AMD can package EPYC CPUs, Instinct GPUs and Pensando networking into a full rack-scale AI platform backed by real demand from $META, $ORCL and OpenAI.

Shay Boloor

89,204 просмотров • 1 месяц назад

🎤 Overheard at #CES2026: AI compute is entering the yottascale era. William Myrhang shows how the AMD Helios AI Platform enables AI to scale by the rack and by the data center, not one server at a time. Designed for agentic and large-scale AI workloads, Helios makes yottascale compute practical, deployable, and sustainable. Zettaflops defined the early AI boom. Yottaflops will define what comes next. AMD is helping build the open foundation needed to support this unprecedented scale.

🎤 Overheard at #CES2026: AI compute is entering the yottascale era. William Myrhang shows how the AMD Helios AI Platform enables AI to scale by the rack and by the data center, not one server at a time. Designed for agentic and large-scale AI workloads, Helios makes yottascale compute practical, deployable, and sustainable. Zettaflops defined the early AI boom. Yottaflops will define what comes next. AMD is helping build the open foundation needed to support this unprecedented scale.

AMD

21,014 просмотров • 7 месяцев назад

“Helios is truly a gamechanger.” Well said, Dr. Lisa Su. This unified rack system, integrating AMD CPUs, GPUs, Pensando NICs, and ROCm software, is purpose-built for the most demanding AI workloads. Helios delivers leadership memory, bandwidth, and interconnect speed. 2026 is not that far away. Get ready! #AdvancingAI

“Helios is truly a gamechanger.” Well said, Dr. Lisa Su. This unified rack system, integrating AMD CPUs, GPUs, Pensando NICs, and ROCm software, is purpose-built for the most demanding AI workloads. Helios delivers leadership memory, bandwidth, and interconnect speed. 2026 is not that far away. Get ready! #AdvancingAI

AMD

60,134 просмотров • 1 год назад

Ayar Labs and Wiwynn are partnering to bring CPO to rack-scale AI systems that support next-generation hyperscale AI workloads. By combining Ayar Labs’ CPO solution with Wiwynn’s rack-level system design and manufacturing capabilities, the two companies are enabling a new class of rack-scale AI infrastructure.

Ayar Labs and Wiwynn are partnering to bring CPO to rack-scale AI systems that support next-generation hyperscale AI workloads. By combining Ayar Labs’ CPO solution with Wiwynn’s rack-level system design and manufacturing capabilities, the two companies are enabling a new class of rack-scale AI infrastructure.

Ayar Labs

137,455 просмотров • 3 месяцев назад

$NVDA is evolving into the platform behind the entire AI factory spanning compute, CPUs, networking, interconnect, software, rack-scale systems and now the financing that helps customers build them. Owning more of that stack lets Nvidia capture more of every dollar spent on AI infrastructure while sitting across the bottlenecks that determine how quickly the entire market can scale.

$NVDA is evolving into the platform behind the entire AI factory spanning compute, CPUs, networking, interconnect, software, rack-scale systems and now the financing that helps customers build them. Owning more of that stack lets Nvidia capture more of every dollar spent on AI infrastructure while sitting across the bottlenecks that determine how quickly the entire market can scale.

Shay Boloor

153,357 просмотров • 11 дней назад

10x more tokens/second per megawatt as measured by CoreWeave on NVIDIA Vera Rubin platform. That performance reflects a bigger shift in AI infrastructure economics: building AI factories that efficiently turn compute into tokens, and tokens into revenue. It takes extreme co-design across the full stack, combined with an open ecosystem built to innovate and scale. Dion Harris connects the economics to the technology with 5 critical questions every AI infrastructure leader should be asking across compute, networking, storage, software and security. ➡️

10x more tokens/second per megawatt as measured by CoreWeave on NVIDIA Vera Rubin platform. That performance reflects a bigger shift in AI infrastructure economics: building AI factories that efficiently turn compute into tokens, and tokens into revenue. It takes extreme co-design across the full stack, combined with an open ecosystem built to innovate and scale. Dion Harris connects the economics to the technology with 5 critical questions every AI infrastructure leader should be asking across compute, networking, storage, software and security. ➡️

NVIDIA AI Infrastructure

23,393 просмотров • 16 дней назад

Vulcano 🌋, our next-generation scale-out AI NIC, is ready to erupt with performance. Fully UC 1.0 compliant; delivers 800 Gbps line rate throughput and directly connects to CPUs and GPUs via PCIe and UL interfaces. Engineered to scale the largest AI systems. Get ready for unparalleled AI networking. #AdvancingAI

Vulcano 🌋, our next-generation scale-out AI NIC, is ready to erupt with performance. Fully UC 1.0 compliant; delivers 800 Gbps line rate throughput and directly connects to CPUs and GPUs via PCIe and UL interfaces. Engineered to scale the largest AI systems. Get ready for unparalleled AI networking. #AdvancingAI

AMD

29,840 просмотров • 1 год назад

The future of compute is open. Vultr and AMD are delivering a collaborative, cloud-native ecosystem for next-gen enterprise applications. 📺 See why delivering AI at scale takes more than GPUs.

The future of compute is open. Vultr and AMD are delivering a collaborative, cloud-native ecosystem for next-gen enterprise applications. 📺 See why delivering AI at scale takes more than GPUs.

AMD

26,391 просмотров • 10 месяцев назад

From training to inference at massive scale, delivering on the promise of AI depends on collaboration across open hardware and software. AMD and OpenAI are working together to tailor hardware like Instinct MI455X and Helios for where AI workloads are going, not just where they are today. Greg Brockman (Greg Brockman) on the power of our engineering partnership at #CES2026.

From training to inference at massive scale, delivering on the promise of AI depends on collaboration across open hardware and software. AMD and OpenAI are working together to tailor hardware like Instinct MI455X and Helios for where AI workloads are going, not just where they are today. Greg Brockman (Greg Brockman) on the power of our engineering partnership at #CES2026.

AMD

20,021 просмотров • 7 месяцев назад

$MU $SKHY $DRAM Holy Shit $AMD's Lisa Su just spent two full minutes at the 2026 AI keynote raving about HBM4 capacity as the thing that separates Helios from the competition. Not compute. Memory. "When you compare Helios to the competition, we're delivering 15% more compute, 50% more HBM4 memory capacity and memory bandwidth, and 50% more scale-out bandwidth." Notice which number is bigger. "What that means is that every Helios can deliver more performance for the largest models, more capacity for longer context, and the bandwidth to scale across thousands of racks.... More than 18,000 CDNA five GPU compute units, over 4,600 Zen 6 CPU cores, and 31 terabytes of HBM4 memory, all in a single rack." "That's what it takes to run agentic AI at scale. Today, I'm excited to announce that Helios is in full production. We have shipments on track to start at the end of the third quarter and ramping into the fourth quarter in the second half of the year. I can tell you, customer demand for Helios is extremely strong, we're extremely proud of the work that we have done across the leading AI labs to adopt Helios."

$MU $SKHY $DRAM Holy Shit $AMD's Lisa Su just spent two full minutes at the 2026 AI keynote raving about HBM4 capacity as the thing that separates Helios from the competition. Not compute. Memory. "When you compare Helios to the competition, we're delivering 15% more compute, 50% more HBM4 memory capacity and memory bandwidth, and 50% more scale-out bandwidth." Notice which number is bigger. "What that means is that every Helios can deliver more performance for the largest models, more capacity for longer context, and the bandwidth to scale across thousands of racks.... More than 18,000 CDNA five GPU compute units, over 4,600 Zen 6 CPU cores, and 31 terabytes of HBM4 memory, all in a single rack." "That's what it takes to run agentic AI at scale. Today, I'm excited to announce that Helios is in full production. We have shipments on track to start at the end of the third quarter and ramping into the fourth quarter in the second half of the year. I can tell you, customer demand for Helios is extremely strong, we're extremely proud of the work that we have done across the leading AI labs to adopt Helios."

Trade Whisperer

180,418 просмотров • 1 месяц назад

Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, anywhere, on fit-for-purpose accelerators and network paths, with maximum performance and efficiency. AI workloads have evolved beyond large-scale pre-training. Today, they encompass fine-tuning, reinforcement learning (RL), synthetic data generation, evaluation pipelines, and more. Fairwater is built to support this full lifecycle: Max density: Fairwater’s two-story design and liquid cooling system lets us place racks in three dimensions and pack them with GPUs as densely as possible, minimizing cable runs and improving latency and effective bandwidth. Fleet: Each Fairwater DC can integrate hundreds of thousands of the latest NVIDIA GPUs into a single coherent cluster. This provides flexible infra that can support the full spectrum of workloads, and ensure no GPU is left unnecessarily idle. And that’s on top of the more than 100,000 GB300s coming online this quarter alone for inference across the rest of our fleet. For us, it’s all about turning every gigawatt into the maximum number of useful tokens. Not every GW is created equal! Planet-scale: Every Fairwater DC will connect through our continent-spanning AI WAN to prior generations of AI supercomputers, forming a truly fungible pool of compute. This enables developers to scale beyond the capacity of a single site and dynamically land workloads on the right infra for their needs. Together, these innovations let us bring together different generations of silicon and AI systems across DCs and geos into a single elastic system that scales seamlessly across training and inference workloads And this elastic AI capacity is all available alongside all the other cloud services (compute, storage, databases, app services) that AI agents and workloads need. This is what we mean when we talk about building a fungible fleet – a single, unified platform that pushes the limits of performance per watt and per dollar. Read more:

Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, anywhere, on fit-for-purpose accelerators and network paths, with maximum performance and efficiency. AI workloads have evolved beyond large-scale pre-training. Today, they encompass fine-tuning, reinforcement learning (RL), synthetic data generation, evaluation pipelines, and more. Fairwater is built to support this full lifecycle: Max density: Fairwater’s two-story design and liquid cooling system lets us place racks in three dimensions and pack them with GPUs as densely as possible, minimizing cable runs and improving latency and effective bandwidth. Fleet: Each Fairwater DC can integrate hundreds of thousands of the latest NVIDIA GPUs into a single coherent cluster. This provides flexible infra that can support the full spectrum of workloads, and ensure no GPU is left unnecessarily idle. And that’s on top of the more than 100,000 GB300s coming online this quarter alone for inference across the rest of our fleet. For us, it’s all about turning every gigawatt into the maximum number of useful tokens. Not every GW is created equal! Planet-scale: Every Fairwater DC will connect through our continent-spanning AI WAN to prior generations of AI supercomputers, forming a truly fungible pool of compute. This enables developers to scale beyond the capacity of a single site and dynamically land workloads on the right infra for their needs. Together, these innovations let us bring together different generations of silicon and AI systems across DCs and geos into a single elastic system that scales seamlessly across training and inference workloads And this elastic AI capacity is all available alongside all the other cloud services (compute, storage, databases, app services) that AI agents and workloads need. This is what we mean when we talk about building a fungible fleet – a single, unified platform that pushes the limits of performance per watt and per dollar. Read more:

Satya Nadella

908,065 просмотров • 9 месяцев назад

Today, AMD and Microsoft are announcing an expanded strategic partnership to deliver a full stack of AMD AI solutions for Microsoft Azure, spanning GPUs, CPUs, networking and software. At the center of this expansion, Microsoft will deploy AMD Helios to power frontier model AI inference for Microsoft, its AI customers and support Azure AI services. Read more on the news ahead of #AdvancingAI:

Today, AMD and Microsoft are announcing an expanded strategic partnership to deliver a full stack of AMD AI solutions for Microsoft Azure, spanning GPUs, CPUs, networking and software. At the center of this expansion, Microsoft will deploy AMD Helios to power frontier model AI inference for Microsoft, its AI customers and support Azure AI services. Read more on the news ahead of #AdvancingAI:

AMD

156,284 просмотров • 1 месяц назад

70% more compute performance, up to 256 high-performance Zen 6 cores, built on TSMC's 2nm process. Meet Venice, our next-generation EPYC processor. It's engineered for more performance, better efficiency, and outstanding TCO. The CPU's role in AI is vital, and Venice is ready. #AdvancingAI

70% more compute performance, up to 256 high-performance Zen 6 cores, built on TSMC's 2nm process. Meet Venice, our next-generation EPYC processor. It's engineered for more performance, better efficiency, and outstanding TCO. The CPU's role in AI is vital, and Venice is ready. #AdvancingAI

AMD

46,825 просмотров • 1 год назад

The NVIDIA Vera Rubin platform is a multi-rack pod-scale system, built to process agentic AI and is now in full production. Built through extreme co-design, Vera Rubin unifies five connected rack-scale systems — the NVIDIA Vera Rubin NVL72, Vera CPU rack, Groq 3 LPX, Vera BlueField-4 STX, and Spectrum-6 SPX Ethernet — all purpose-built for global-scale agentic AI infrastructure. Vera Rubin delivers intelligence at scale, with the performance, efficiency and security needed to power the next industrial revolution. Congratulations to Microsoft, Dell Technologies, and CoreWeave for each standing up their Vera Rubin NVL72 engineering racks. Learn more:

The NVIDIA Vera Rubin platform is a multi-rack pod-scale system, built to process agentic AI and is now in full production. Built through extreme co-design, Vera Rubin unifies five connected rack-scale systems — the NVIDIA Vera Rubin NVL72, Vera CPU rack, Groq 3 LPX, Vera BlueField-4 STX, and Spectrum-6 SPX Ethernet — all purpose-built for global-scale agentic AI infrastructure. Vera Rubin delivers intelligence at scale, with the performance, efficiency and security needed to power the next industrial revolution. Congratulations to Microsoft, Dell Technologies, and CoreWeave for each standing up their Vera Rubin NVL72 engineering racks. Learn more:

NVIDIA

49,740 просмотров • 3 месяцев назад

$QCOM buying Modular was about more than adding another AI asset. I sat down with Modular CEO Chris Lattner and Qualcomm CEO Cristiano R. Amon on why the industry needs a better software model for heterogeneous compute and how Qualcomm + Modular intend to attack it. If they can make it dramatically easier for developers to target CPUs, GPUs and accelerators without being locked into a single architecture, this could matter a lot.

$QCOM buying Modular was about more than adding another AI asset. I sat down with Modular CEO Chris Lattner and Qualcomm CEO Cristiano R. Amon on why the industry needs a better software model for heterogeneous compute and how Qualcomm + Modular intend to attack it. If they can make it dramatically easier for developers to target CPUs, GPUs and accelerators without being locked into a single architecture, this could matter a lot.

Patrick Moorhead

54,516 просмотров • 11 дней назад

“This is the most advanced accelerator we've ever built!” - Dr. Lisa Su. Helios' "heart": The MI400 Series. It runs trillion-plus parameter models, delivers up to 40 PFLOPS of FP4 performance, features 432 GB of HBM4, and supports 300 GB/s of scale-out bandwidth to connect across racks and clusters. Now, this is how we power the future of AI. #AdvancingAI

“This is the most advanced accelerator we've ever built!” - Dr. Lisa Su. Helios' "heart": The MI400 Series. It runs trillion-plus parameter models, delivers up to 40 PFLOPS of FP4 performance, features 432 GB of HBM4, and supports 300 GB/s of scale-out bandwidth to connect across racks and clusters. Now, this is how we power the future of AI. #AdvancingAI

AMD

68,513 просмотров • 1 год назад

Moonshot AI limiting Kimi K3 subscriptions because of compute constraints is a massive signal because the 2.8 trillion parameter model is already trained and the bottleneck is now emerging as usage scales. Serving that demand requires a ton more accelerators, enormous memory capacity and faster networking to keep data moving efficiently across the cluster. So as AI spending shifts toward serving millions or billions of daily requests then memory and networking become critical bottlenecks giving $MU, $SKHY, $AVGO, $CRDO, $ALAB and $MRVL direct exposure to inference at scale.

Moonshot AI limiting Kimi K3 subscriptions because of compute constraints is a massive signal because the 2.8 trillion parameter model is already trained and the bottleneck is now emerging as usage scales. Serving that demand requires a ton more accelerators, enormous memory capacity and faster networking to keep data moving efficiently across the cluster. So as AI spending shifts toward serving millions or billions of daily requests then memory and networking become critical bottlenecks giving $MU, $SKHY, $AVGO, $CRDO, $ALAB and $MRVL direct exposure to inference at scale.

Shay Boloor

123,791 просмотров • 1 месяц назад

🎤 Overheard at #CES2026: AMD EVP Phil Guido shares how Helios addresses the performance, cost and compute demands for customers in a single solution.

🎤 Overheard at #CES2026: AMD EVP Phil Guido shares how Helios addresses the performance, cost and compute demands for customers in a single solution.

AMD

19,335 просмотров • 7 месяцев назад