Speaking at an event in San Francisco, AMD CEO Lisa Su unveiled AI-infused chips across the corporate’s Ryzen, Instinct and Epyc brands, fueling a brand new generation of AI computing for everybody from business users to data centers.
Throughout the event, AMD not directly made references to rivals resembling Nvidia and Intel by emphasizing its quest to offer technology that was open and accessible to the widest variety of shoppers, without an intent to lock those customers into proprietary solutions.
Su said AI will boost our personal productivity, collaboration will develop into a lot better with things like real-time translate, and it is going to make life easier whether you might be a creator or atypical user. It will likely be processed locally, to guard your privacy, Su said. She noted the brand new AMD Ryzen AI Pro PCs will likely be CoPilot+-ready and offer as much as 23 hours of battery life (and nine hours using Microsoft Teams).
“We’ve been working very closely with AI PC ecosystem developers,” she said, noting greater than 100 will likely be working on AI apps by the tip of the yr.
Commercial AI mobile Ryzen processors
AMD announced its third generation business AI mobile processors, designed specifically to remodel business productivity with Copilot+ features including live captioning and language translation in conference calls and advanced AI image generators. If you actually desired to, you can use AI-based Microsoft Teams for as much as nine hours on latest laptops equipped with the AMD processors.
The latest Ryzen AI PRO 300 Series processors deliver industry-leading AI compute, with as much as 3 times the AI performance than the previous generation of AMD processors. More than 100 products using the Ryzen processors are on the best way through 2025.
Enabled with AMD PRO Technologies, the Ryzen AI PRO 300 Series processors offer high security and manageability features designed to streamline IT operations and ensure exceptional ROI for businesses.
Ryzen AI PRO 300 Series processors feature latest AMD Zen 5 architecture, delivering outstanding CPU performance, and are the world’s best line up of economic processors for Copilot+ enterprise PCs5. Zen, now in its fifth generation, has been the muse behind AMD’s own financial recovery, its gains in market share against Intel, and Intel’s own subsequent hard times and layoffs.
“I feel the most effective is that AMD proceed to execute on a solid product roadmap. Unfortunately they’re making performance comparisons to the competition’s previous generation products,” said Jim McGregor, an analyst at Tirias Research, in an email to VentureBeat. “So, now we have to attend and see how the products will compare. However, I do expect them to be highly competitive especially the processors. Note that AMD only announced a brand new architecture for nenetworking, every part else is evolutionary but that’s not a foul thing when you find yourself in a powerful position and gaining market share.”
Laptops equipped with Ryzen AI PRO 300 Series processors are designed to tackle business’ hardest workloads, with the top-of-stack Ryzen AI 9 HX PRO 375 offering as much as 40% higher performance and as much as 14% faster productivity performance in comparison with Intel’s Core Ultra 7 165U, AMD said.
With the addition of XDNA 2 architecture powering the integrated NPU (the neural processing unit, or AI-focused a part of the processor), AMD Ryzen AI PRO 300 Series processors offer a cutting-edge 50+ NPU TOPS (Trillions of Operations Per Second) of AI processing power, exceeding Microsoft’s Copilot+ AI PC requirements and delivering exceptional AI compute and productivity capabilities for the fashionable business.
Built on a 4 nanometer (nm) process and with progressive power management, the brand new processors deliver prolonged battery life ideal for sustained performance and productivity on the go.
“Enterprises are increasingly demanding more compute power and efficiency to drive their on a regular basis tasks and most taxing workloads. We are excited so as to add the Ryzen AI PRO 300 Series, probably the most powerful AI processor built for business PCs10 , to our portfolio of mobile processors,” said Jack Huynh, senior vp and general manager of the computing and graphics group at AMD, in an announcement. “Our third generation AI-enabled processors for business PCs deliver unprecedented AI processing capabilities with incredible battery life and seamless compatibility for the applications users rely upon.”
AMD expands business OEM ecosystem
OEM partners proceed to expand their business offerings with latest PCs powered by Ryzen AI PRO 300 Series processors, delivering well-rounded performance and compatibility to their business customers. With industry leading TOPS, the subsequent generation of Ryzen processor-powered business PCs are set to expand the probabilities of local AI processing with Microsoft Copilot+. OEM systems powered by Ryzen AI PRO 300 Series are expected to be on shelf starting later this yr.
“Microsoft’s partnership with AMD and the combination of Ryzen AI PRO processors into Copilot+ PCs display our joint concentrate on delivering impactful AI-driven experiences for our customers. The Ryzen AI PRO’s performance, combined with the most recent features in Windows 11, enhances productivity, efficiency, and security,” said Pavan Davuluri, corporate vp for Windows+ Devices at Microsoft, in an announcement. “Features like Improved Windows Search, Recall, and Click to Do make PCs more intuitive and responsive. Security enhancements, including the Microsoft Pluton security processor and Windows Hello Enhanced Sign-in Security, help safeguard customer data with advanced protection. We’re pleased with our strong history of collaboration with AMD and are thrilled to bring these innovations to market.”
“In today’s AI-powered era of computing, HP is devoted to delivering powerful innovation and performance that revolutionizes the best way people work,” said Alex Cho, president of Personal Systems at HP, in an announcement. “With the HP EliteBook X Next-Gen AI PC, we’re empowering modern leaders to push boundaries without compromising power or performance. We are proud to expand our AI PC lineup powered by AMD, providing our business customers with a very personalized experience.”
“Lenovo’s partnership with AMD continues to drive AI PC innovation and deliver supreme performance for our business customers. Our recently announced ThinkPad T14s Gen 6 AMD, powered by the most recent AMD Ryzen AI PRO 300 Series processors, showcases the strength of our collaboration,” said Luca Rossi, president, Lenovo Intelligent Devices Group. “This device offers outstanding AI computing power, enhanced security, and exceptional battery life, providing professionals with the tools they should maximize productivity and efficiency. Together with AMD, we’re transforming the business landscape by delivering smarter, AIdriven solutions that empower users to realize more.”
New Pro Technologies features for security and management
In addition to AMD Secure Processor, AMD Shadow Stack and AMD Platform Secure Boot, AMD has expanded its Pro Technologies lineup with latest security and manageability features.
Processors equipped with PRO Technologies will now come standard with Cloud Bare Metal Recovery, allowing IT teams to seamlessly recuperate systems via the cloud ensuring smooth and continuous operations; Supply Chain Security (AMD Device Identity), a brand new supply chain security function, enabling traceability across the provision chain; and Watch Dog Timer, constructing on existing resiliency support with additional detection and recovery processes.
Additional AI-based malware detection is on the market via PRO Technologies with select ISV partners. These latest safety features leverage the integrated NPU to run AI-based security workloads without impacting day-to-day performance.
AMD unveils Instinct MI325X accelerators for AI data centers
AMD has develop into a giant player within the graphics processing units (GPUs) for data centers, and today it announced the most recent AI accelerators and networking solutions for AI infrastructure.
The company unveiled the AMD Instinct MI325X accelerators, the AMD Pensando Pollara 400
network interface card (NIC) and the AMD Pensando Salina data processing unit (DPU).
AMD claimed the AMD Instinct MI325X accelerators set a brand new standard in performance for Gen AI models and data centers. Built on the AMD CDNA 3 architecture, AMD Instinct MI325X accelerators are designed for performance and efficiency for demanding AI tasks spanning foundation model training, fine-tuning and inferencing.
Together, these products enable AMD customers and partners to create highly performant and optimized AI solutions on the system, rack and data center level.
“AMD continues to deliver on our roadmap, offering customers the performance they need and the alternative they need, to bring AI infrastructure, at scale, to market faster,” said Forrest Norrod, executive vp and general manager of the info center solutions business group at AMD, in an announcement. “With the brand new AMD Instinct accelerators, EPYC processors and AMD Pensando networking engines, the continued growth of our open software ecosystem, and the flexibility to tie this all together into optimized AI infrastructure, AMD underscores the critical expertise to construct and deploy world class AI solutions.”
AMD Instinct MI325X accelerators deliver industry-leading memory capability and bandwidth, with 256GB of HBM3E supporting 6.0TB/s offering 1.8 times more capability and 1.3 times more bandwidth than the Nvidia H200, AMD said. The AMD Instinct MI325X also offers 1.3 times greater peak theoretical FP16 and FP8 compute performance in comparison with H200.
This leadership memory and compute can provide as much as 1.3 times the inference performance on Mistral 7B at FP162, 1.2 times the inference performance on Llama 3.1 70B at FP83 and 1.4 times the inference performance on Mixtral 8x7B at FP16 of the H200. (Nvidia has newer devices available on the market now and so they usually are not yet available for comparisons, AMD said).
“AMD actually stays well positioned in the info center, but I feel their CPU efforts are still their best positioned products. The marketplace for AI accelleration/GPUs continues to be heavily favoring Nvidia and I don’t see that changing anytime soon. But the necessity for well optimized and purpose designed CPUs to go with as a bunch processor any AI accelerator or GPU is important and AMDs datacenter CPUs are competitive there,” said Ben Bajarin, an analyst at Creative Strategies, in an email to VentureBeat. “On the networking front, there’s actually good progress here technically and I imagine the more AMD can integrate this into their full stack approach to optimizing for the racks via the ZT systems purchase, then I feel their networking stuff becomes much more essential.”
He added, “Broad point to make here, is the info center is under a whole transformation and we’re still only within the early days of that which makes this still a large open competitive field over the arc of time 10+ years. I’m undecided we will say with any certainty how this shakes out over that point but the underside line is there’s a number of market share and $$ to go around to maintain AMD, Nvidia, and Intel busy.”
AMD Instinct MI325X accelerators are currently on course for production shipments in Q4 2024 and are expected to have widespread system availability from a broad set of platform providers, including Dell Technologies, Eviden, Gigabyte, Hewlett Packard Enterprise, Lenovo, Supermicro and others starting in Q1 2025.
Updating its annual roadmap, AMD previewed the next-generation AMD Instinct MI350 series accelerators. Based on the AMD CDNA 4 architecture, AMD Instinct MI350 series accelerators are designed to deliver a 35 times improvement in inference performance in comparison with AMD CDNA 3-based accelerators.
The AMD Instinct MI350 series will proceed to drive memory capability leadership with as much as 288GB of HBM3E memory per accelerator. The AMD Instinct MI350 series accelerators are on course to be available throughout the second half of 2025.
“AMD undoubtedly increased the gap between itself and Intel with Epyc. It currently has 50-60% market share with the hyoerscalers and I don’t see that abating. AMD;’s biggest challenge is to get share with enterprises. Best product rarely wins within the enterprise and AMD needs to speculate more into sales and marketing to speed up its enterprise growth,” said Patrick Moorhead, an analyst at Moor Insights & Strategy, in an email to VentureBeat. “It’s s bit harder to evaluate where AMD sits versus NVIDIA in Datacenter GPUs. There’s numbers flying throughout, claims from each firms that they’re higher. Signal65, our sister benchmarking company, hasn’t had the chance to do our own tests.”
And Moohead added, “What I can unequivocally say is that AMD’s latest GPUs, particularly the MI350, is an enormous improvement given improved efficiency, performance and higher support for lower bit rate models than its predecessors. It is a two horse race, with Nvidia in the large lead and AMD is quickly catching up and providing meaningful results. The facts that Meta’s live llama 405B model runs exclusively on MI is a large statement on competitiveness. “
AMD next-gen AI Networking
AMD is leveraging probably the most widely deployed programmable DPU for hyperscalers to power next-gen AI networking, said Soni Jiandani, senior vp of the network technology solutions group, in a press briefing.
Split into two parts: the front-end, which delivers data and knowledge to an AI cluster, and the backend, which manages data transfer between accelerators and clusters, AI networking is critical to making sure CPUs and accelerators are utilized efficiently in AI infrastructure.
To effectively manage these two networks and drive high performance, scalability and efficiency across your entire system, AMD introduced the AMD Pensando Salina DPU for the front-end and the AMD Pensando Pollara 400, the industry’s first Ultra Ethernet Consortium (UEC) ready AI NIC, for the back-end.
The AMD Pensando Salina DPU is the third generation of the world’s most performant and programmable DPU, bringing as much as two times the performance, bandwidth and scale in comparison with the previous generation.
Supporting 400G throughput for fast data transfer rates, the AMD Pensando Salina DPU is a critical component in AI front-end network clusters, optimizing performance, efficiency, security and scalability for data-driven AI applications.
The UEC-ready AMD Pensando Pollara 400, powered by the AMD P4 Programmable engine, is the industry’s first UEC-ready AI NIC. It supports the next-gen RDMA software and is backed by an open ecosystem of networking. The AMD Pensando Pollara 400 is critical for providing leadership performance, scalability and efficiency of accelerator-to-accelerator communication in back-end networks.
Both the AMD Pensando Salina DPU and AMD Pensando Pollara 400 are sampling with customers in Q4’24 and are on course for availability in the primary half of 2025.
AMD AI software for Generative AI
AMD continues its investment in driving software capabilities and the open ecosystem to deliver powerful latest features and capabilities within the AMD ROCm open software stack.
Within the open software community, AMD is driving support for AMD compute engines in probably the most widely used AI frameworks, libraries and models including PyTorch, Triton, Hugging Face and lots of others. This work translates to out-of-the-box performance and support with AMD Instinct accelerators on popular generative AI models like Stable Diffusion 3, Meta Llama 3, 3.1 and three.2 and multiple million models at Hugging Face.
Beyond the community, AMD continues to advance its ROCm open software stack, bringing the most recent features to support leading training and inference on Generative AI workloads. ROCm 6.2 now includes support for critical AI features like FP8 datatype, Flash Attention 3, Kernel Fusion and more. With these latest additions, ROCm 6.2, in comparison with ROCm 6.0, provides as much as a 2.4X performance improvement on inference6 and 1.8X on training for quite a lot of LLMs.
AMD launches fifth Gen AMD Epyc CPUs for the info center
AMD also announced the supply of the fifth Gen AMD Epyc processors, formerly codenamed “Turin,” the “world’s best server CPU for enterprise, AI and cloud,” the corporate said.
Using the Zen 5 core architecture, compatible with the broadly deployed SP5 platform and offering a broad range of core counts spanning from eight to 192, the AMD Epyc 9005 Series processors extend the record-breaking performance and energy efficiency of the previous generations with the highest of stack 192 core CPU delivering as much as 2.7 times the performance in comparison with the competition, AMD said.
New to the AMD Epyc 9005 Series CPUs is the 64 core AMD Epyc 9575F, tailor made for GPU-powered AI solutions that need the last word in host CPU capabilities. Boosting as much as 5GHz, in comparison with the three.8GHz processor of the competition, it provides up to twenty-eight% faster processing needed to maintain GPUs fed with data for demanding AI workloads, AMD said.
“From powering the world’s fastest supercomputers, to leading enterprises, to the biggest Hyperscalers, AMD has earned the trust of shoppers who value demonstrated performance, innovation and energy efficiency,” said Dan McNamara, senior vp and general manager of the server business at AMD, in an announcement. “With five generations of on-time roadmap execution, AMD has proven it may possibly meet the needs of the info center market and provides customers the usual for data center performance, efficiency, solutions and capabilities for cloud, enterprise and AI workloads.”
In a press briefing, McNamara thanked Zen for AMD’s server market share rise from zero in 2017 to 34% within the second quarter of 2024 (in accordance with Mercury Research).
Modern data centers run quite a lot of workloads, from supporting corporate AI-enablement initiatives, to powering large-scale cloud-based infrastructures to hosting probably the most demanding business-critical applications. The latest fifth Gen AMD Epyc processors provide leading performance and capabilities for the broad spectrum of server workloads driving business IT today.
“This is a beast,” McNamara said. “We are really enthusiastic about it.”
The latest Zen 5 core architecture, provides as much as 17% higher instructions per clock (IPC) for enterprise and cloud workloads and as much as 37% higher IPC in AI and high performance computing (HPC) in comparison with Zen 4.
With AMD Epyc 9965 processor-based servers, customers can expect significant impact of their real world applications and workloads in comparison with the Intel Xeon 8592+ CPU-based servers, with: as much as 4 times faster time to results on business applications resembling video transcoding.
AMD said it also has as much as 3.9 times the time to insights for science and HPC applications that solve the
world’s most difficult problems; as much as 1.6 times the performance per core in virtualized infrastructure.
In addition to leadership performance and efficiency usually purpose workloads, the fifth Gen
AMD Epyc processors enable customers to drive fast time to insights and deployments for AI
deployments, whether or not they are running a CPU or a CPU + GPU solution, McNamara said.
Compared to the competition, he said the 192 core Epyc 9965 CPU has as much as 3.7 times the performance on end-to-end AI workloads, like TPCx-AI (derivative), that are critical for driving an efficient approach to generative AI.
In small and medium size enterprise-class generative AI models, like Meta’s Llama 3.1-8B, the Epyc 9965 provides 1.9 times the throughput performance in comparison with the competition.
Finally, the aim built AI host node CPU, the EPYC 9575F, can use its 5GHz max frequency boost to assist a 1,000 node AI cluster drive as much as 700,000 more inference tokens per second. Accomplishing more, faster.
By modernizing to an information center powered by these latest processors to realize 391,000 units of SPECrate2017_int_base general purpose computing performance, customers receive impressive performance for various workloads, while gaining the flexibility to make use of an estimated 71% less power and ~87% fewer servers. This gives CIOs the pliability to either profit from the space and power savings or add performance for day-to-day IT tasks while delivering impressive AI performance.
The entire lineup of fifth Gen AMD EPYC processors is on the market today, with support from Cisco, Dell, Hewlett Packard Enterprise, Lenovo and Supermicro in addition to all major ODMs and cloud service providers providing a straightforward upgrade path for organizations in search of compute and AI leadership.
Dell said that said its 16-accelerated PowerEdge servers would give you the option to exchange seven prior generation servers, with a 65% reduction of energy usage. HP Enterprise also took the stage to say Lumi, certainly one of its customers, is working on a digital twin of your entire planet, dubbed Destination Earth, using the AMD tech.
Daniel Newman, CEO of The Futurum Group and an analyst, said in an email to VentureBeat, “Instinct and the brand new MI325X will likely be the recent button from today’s event. It isn’t a very latest launch, however the Q4 ramp will run alongside nvidia Blackwell and will likely be the subsequent essential indicator of AMD’s trajectory as probably the most compelling competitor to Nvidia. The 325X ramping while the brand new 350 will likely be the largest leap when it launches in 2H of 2025 making a 35 times AI performance leap from its CDNA3. “
Newman added, “Lisa Su’s declaration of a $500 billion AI accelerator market between 2023 and 2028 is an incredibly ambitious leap that represents greater than 2x our current forecast and indicates a fabric upside for the market coming from a typically conservative CEO in Lisa Su. Other announcements in networking and compute (Turin) show the corporate’s continued expansion and growth.”
And he said, “The Epyc DC CPU business showed significant generational improvements. AMD has been incredibly successful in winning cloud datacenter business for its EPYC line now having greater than 50% of share and in some cases we consider closer to 80%. For AMD, the large query is can it turn the strength in cloud and switch its attention to enterprise data center where Intel continues to be dominant–this might see AMD DC CPU business expand to greater than its already largest ever 34%. Furthermore, can the corporate reap the benefits of its strength in cloud to win more DC GPU deals and fend off NVIDIA’s strength at greater than 90% market share.”
