The AMD Instinct software ecosystem keeps growing.
@DriveNets brings its AI networking fabric and cluster orchestration to AMD Instinct GPUs, helping operators turn multi-node deployments into a single validated AI cluster with predictable communication performance.
Together, aking large Instinct clusters faster to deploy and easier to operate.
Explore DriveNets and the growing AMD Instinct software ecosystem: amd.com/en/ecosystem/i…
@DriveNets brings its AI networking fabric and cluster orchestration to AMD Instinct GPUs, helping operators turn multi-node deployments into a single validated AI cluster with predictable communication performance.
Together, aking large Instinct clusters faster to deploy and easier to operate.
Explore DriveNets and the growing AMD Instinct software ecosystem: amd.com/en/ecosystem/i…
1
0
7
Agentic AI demands 5 to 30x more tokens per task than a standard query (Gartner).
Yousuf Khan, CVP of NTSG at AMD, explains how the DPU takes on networking, security, and storage work so CPU resources stay focused on AI workloads. Read more: bit.ly/4jpsNgY
Yousuf Khan, CVP of NTSG at AMD, explains how the DPU takes on networking, security, and storage work so CPU resources stay focused on AI workloads. Read more: bit.ly/4jpsNgY
0
1
7
1,536 GPUs. Approximately 20 trillion tokens trained. Training runs lasting up to 72 hours.
Swipe through the results behind @ZyphraAI’s ZAYA1-8B reasoning model, trained from scratch on a full-stack AMD AI platform.
See how AMD Instinct GPUs, AMD ROCm software and AMD king came together at scale: bit.ly/4ixfdaB
Swipe through the results behind @ZyphraAI’s ZAYA1-8B reasoning model, trained from scratch on a full-stack AMD AI platform.
See how AMD Instinct GPUs, AMD ROCm software and AMD king came together at scale: bit.ly/4ixfdaB
1
2
30
Evaluate whether your data center infrastructure is ready to support AI workloads and where modernization can help you scale.
Grounded in AMD's expertise in AI infrastructure, this personalized assessment helps you understand how prepared your data center is to support AI today. rtunities to strengthen your foundation as demands grow.
Take the assessment: readiness.amd.com
Grounded in AMD's expertise in AI infrastructure, this personalized assessment helps you understand how prepared your data center is to support AI today. rtunities to strengthen your foundation as demands grow.
Take the assessment: readiness.amd.com
1
0
5
Inside AMD Helios: Engineering AI at Rack Scale.
Join Andrew Dieckmann, Steve Scott and Alan Smith for a fireside chat at #OCPSummit26.
📅 Oct. 12 | 12:50 p.m. PT
📍 SJCC - Concourse Level, 210AE
View the agenda: 2026ocpglobal.fnvirtual.app/a/schedule/
Join Andrew Dieckmann, Steve Scott and Alan Smith for a fireside chat at #OCPSummit26.
📅 Oct. 12 | 12:50 p.m. PT
📍 SJCC - Concourse Level, 210AE
View the agenda: 2026ocpglobal.fnvirtual.app/a/schedule/
0
1
11
Production AI takes collaboration across the stack.
For @character_ai, AMD Instinct™ GPUs and ROCm™ software were paired with hands-on engineering from AMD and @digitalocean across model kernels, infrastructure and production optimization.
Character․ai put it simply: “This n our serving stack without having to reinvent our approach.”
For @character_ai, AMD Instinct™ GPUs and ROCm™ software were paired with hands-on engineering from AMD and @digitalocean across model kernels, infrastructure and production optimization.
Character․ai put it simply: “This n our serving stack without having to reinvent our approach.”
2
0
10
As AI scales, network performance can directly impact GPU efficiency.
AI networking helps keep workloads synchronized and running smoothly at scale.
Missed our blog “What is AI Networking?” Read it here: amd.com/en/blogs/2026/…
AI networking helps keep workloads synchronized and running smoothly at scale.
Missed our blog “What is AI Networking?” Read it here: amd.com/en/blogs/2026/…
0
0
1
AI infrastructure needs software that evolves alongside the hardware.
Elio Van Puyvelde, CEO and founder of Eliovp, shares how his team uses AMD ROCm software, to optimize workloads across generations of AMD Instinct GPUs.
Hear why software optimization is a critical part of k.
Elio Van Puyvelde, CEO and founder of Eliovp, shares how his team uses AMD ROCm software, to optimize workloads across generations of AMD Instinct GPUs.
Hear why software optimization is a critical part of k.
0
1
6
It’s that moment when technology translates into real outcomes that keeps teams building, learning and pushing what’s possible.
Hear Elio Van Puyvelde, CEO and founder of Eliovp, share the excitement of building and training tailored models on AMD hardware, putting them into ustomers experience the results firsthand.
Hear Elio Van Puyvelde, CEO and founder of Eliovp, share the excitement of building and training tailored models on AMD hardware, putting them into ustomers experience the results firsthand.
0
1
7
AI infrastructure will change. The question is how much has to change with it.
Compute, networking, servers and other layers evolve on different timelines. Designing those layers to evolve independently can give organizations greater flexibility to adapt without rebuilding
Explore how infrastructure designed for flexibility can evolve one layer at a time: bit.ly/4xtDkek
Compute, networking, servers and other layers evolve on different timelines. Designing those layers to evolve independently can give organizations greater flexibility to adapt without rebuilding
Explore how infrastructure designed for flexibility can evolve one layer at a time: bit.ly/4xtDkek
2
3
28
Windows Server 2025 delivers modern security, but many features depend on the hardware beneath it.
AMD EPYC 4005 Server CPUs help small businesses and remote offices/branch offices get the most from their upgrade, with security from silicon to OS. Read the blog to see why. bit.ly/4gwgu0l
AMD EPYC 4005 Server CPUs help small businesses and remote offices/branch offices get the most from their upgrade, with security from silicon to OS. Read the blog to see why. bit.ly/4gwgu0l
0
1
9
AI infrastructure economics start with the cost of compute.
@Signal_65 testing hosted via @tensorwave found AMD Instinct MI355X offered a 39.6% GPU-hour cost advantage based on blended on-demand cloud pricing. bit.ly/45hAvl3
@Signal_65 testing hosted via @tensorwave found AMD Instinct MI355X offered a 39.6% GPU-hour cost advantage based on blended on-demand cloud pricing. bit.ly/45hAvl3
2
2
25
Get more from every AI infrastructure dollar.
In @Signal_65 testing hosted via @tensorwave, AMD Instinct MI355X delivered more tokens per dollar across all three models tested, with up to 2.15x the price-performance. bit.ly/45hAvl3
In @Signal_65 testing hosted via @tensorwave, AMD Instinct MI355X delivered more tokens per dollar across all three models tested, with up to 2.15x the price-performance. bit.ly/45hAvl3
0
1
8
AI inference is about more than peak throughput.
Independent @Signal_65 testing shows AMD Instinct MI355X led 2 of 3 models tested and delivered 1.42×–2.15× tok/$ across all 3, even on Kimi-K2.6, where the competition led median throughput. Testing was hosted via @tensorwave. o/4Ll7XzPi8Y
Independent @Signal_65 testing shows AMD Instinct MI355X led 2 of 3 models tested and delivered 1.42×–2.15× tok/$ across all 3, even on Kimi-K2.6, where the competition led median throughput. Testing was hosted via @tensorwave. o/4Ll7XzPi8Y
0
4
29
Easy to integrate. Open by design.
Alex Carter, CEO and co-founder of XPerf AI, shares how AMD ROCm gives developers access to the tools they need, helping save development time.
Alex Carter, CEO and co-founder of XPerf AI, shares how AMD ROCm gives developers access to the tools they need, helping save development time.
0
1
19
Getting more from AI infrastructure starts with making better use of the hardware you already have.
Hear how XPerf AI works with AMD to boost hardware utilization, reliability, and the return on every GPU.
Hear how XPerf AI works with AMD to boost hardware utilization, reliability, and the return on every GPU.
0
3
15
You can't optimize what you can't see.
As AI infrastructure grows more complex, IDC highlights observability and operational efficiency as critical to maintaining performance and resiliency.
Learn more: bit.ly/4quKrkB
As AI infrastructure grows more complex, IDC highlights observability and operational efficiency as critical to maintaining performance and resiliency.
Learn more: bit.ly/4quKrkB
0
2
13
MetaRoCE is more than a new transport protocol.
It demonstrates how programmable AI networking can help infrastructure evolve at the pace of AI. Developed by Meta and implemented on the AMD Pensando Pollara 400 AI NIC, with plans to be deployed on the AMD Pensando Vulcano 800 rries forward on a shared programmable foundation, without need to start over with each hardware generation.
Read more: amd.com/en/blogs/2026/…
It demonstrates how programmable AI networking can help infrastructure evolve at the pace of AI. Developed by Meta and implemented on the AMD Pensando Pollara 400 AI NIC, with plans to be deployed on the AMD Pensando Vulcano 800 rries forward on a shared programmable foundation, without need to start over with each hardware generation.
Read more: amd.com/en/blogs/2026/…
0
3
25
The most expensive GPU is the one sitting idle.
IDC forecasts the AI scale-out networking market will reach $79B by 2030, reflecting networking's growing role in keeping AI infrastructure running efficiently at scale.
Read the IDC Spotlight: bit.ly/4quKrkB
IDC forecasts the AI scale-out networking market will reach $79B by 2030, reflecting networking's growing role in keeping AI infrastructure running efficiently at scale.
Read the IDC Spotlight: bit.ly/4quKrkB
0
0
9
AI is no longer just a compute problem.
IDC forecasts AI infrastructure spending will grow from $486B in 2025 to $1.38T by 2030, highlighting the growing importance of networking, software, and systems design. Learn more: bit.ly/4quKrkB
IDC forecasts AI infrastructure spending will grow from $486B in 2025 to $1.38T by 2030, highlighting the growing importance of networking, software, and systems design. Learn more: bit.ly/4quKrkB
0
2
11
The network has to evolve as fast as AI does. The AMD Pensando Vulcano 800 AI NIC is fully programmable, built for scale-out and scale-across, keeping every GPU productive at scale.
Read more: bit.ly/4zhajET
Read more: bit.ly/4zhajET
7
8
76
Agentic AI demands more from CPUs. AMD EPYC 9006 Series Server CPUs offer:
✅ Up to 512 threads for agent execution
✅ Up to 1.6 TB/s memory bandwidth and up to 5 GHz frequency to help keep GPUs utilized
✅ Leadership performance and performance per watt for general-purpose bout AMD EPYC 9006 Series Server CPUs for agentic AI: bit.ly/4xrkp4d
✅ Up to 512 threads for agent execution
✅ Up to 1.6 TB/s memory bandwidth and up to 5 GHz frequency to help keep GPUs utilized
✅ Leadership performance and performance per watt for general-purpose bout AMD EPYC 9006 Series Server CPUs for agentic AI: bit.ly/4xrkp4d
0
10
49
What makes AMD EPYC 9006 Series Server CPUs the world’s best server CPU for cloud, enterprise, and AI?
✅ Leadership performance and performance per watt
✅ Four purpose-built CPU families
✅ Agent sandbox, AI host node, and general-purpose CPUs for agentic AI
See 5 reasons to ries Server CPUs: bit.ly/4xrkp4d
✅ Leadership performance and performance per watt
✅ Four purpose-built CPU families
✅ Agent sandbox, AI host node, and general-purpose CPUs for agentic AI
See 5 reasons to ries Server CPUs: bit.ly/4xrkp4d
0
6
35
AMD Helios brings CPUs, GPUs and networking together into one rackscale AI platform.
Built for AI at scale.
Built for AI at scale.
25
85
725
Agentic AI requires more than one CPU profile.
The AMD EPYC 9006 Series Server CPU portfolio is purpose-built to scale agent sandboxes, feed accelerators and power enterprise workloads, helping match the right CPU to every stage of the workflow.
Learn more: bit.ly/3TtkTbn
The AMD EPYC 9006 Series Server CPU portfolio is purpose-built to scale agent sandboxes, feed accelerators and power enterprise workloads, helping match the right CPU to every stage of the workflow.
Learn more: bit.ly/3TtkTbn
0
4
44
Introducing AMD Helios.
A rackscale AI platform built to help organizations deploy frontier AI at scale.
A rackscale AI platform built to help organizations deploy frontier AI at scale.
1
15
132
Compute leadership. Open platforms. Powering AI everywhere.
Hear Dr. @LisaSu share the strategy guiding the future of AI at AMD.
Hear Dr. @LisaSu share the strategy guiding the future of AI at AMD.
3
14
118
Agentic AI changes more than software. It changes infrastructure.
Learn why AI agents are driving the next wave of compute demand.
Learn why AI agents are driving the next wave of compute demand.
1
4
44
Agentic AI is changing how data centers are built.
AI agents spend much of their time planning, retrieving data, executing code and orchestrating workflows, making the CPU foundational to AI infrastructure.
Learn how AMD EPYC 9006 Series Server CPUs are built for the agentic TzukV
AI agents spend much of their time planning, retrieving data, executing code and orchestrating workflows, making the CPU foundational to AI infrastructure.
Learn how AMD EPYC 9006 Series Server CPUs are built for the agentic TzukV
1
1
33
AI is transforming every industry, and we're still in the early innings.
Hear Dr. @LisaSu share why AI is driving the next era of computing.
Hear Dr. @LisaSu share why AI is driving the next era of computing.
2
4
48
Independent @Signal_65 testing shows AMD Instinct MI355X delivers up to 1.3× higher throughput, 2× more users at SLA and leading inference economics vs. the competition.
Read the report: bit.ly/45hAvl3
Read the report: bit.ly/45hAvl3
1
2
35
Agentic AI is reshaping AI infrastructure.
As AI workflows become more complex, CPUs are essential for orchestrating the tasks that surround the model. Learn why high thread density, memory performance and rackscale design matter for AI built to scale.
As AI workflows become more complex, CPUs are essential for orchestrating the tasks that surround the model. Learn why high thread density, memory performance and rackscale design matter for AI built to scale.
1
1
25
Enterprise-grade AI visual design, powered by AMD Instinct.
Ideogram just released Ideogram 4.0, a 9.3B-parameter open-weight Diffusion Transformer that outperforms all open-weight image models in LMArena and Design Arena, with a 0.97 X-Omni English OCR text-rendering score (vs. 0.30–0.50).
The engine behind it? AMD Instinct MI350X GPUs, 288 GB of HBM3 memory and 8 TB/s of bandwidth, delivering the throughput enterprise diffusion workloads demand.
Read how @ideogram_ai and AMD are scaling production-grade visual AI from millions to billions of images per month 👉 bit.ly/3TosU1m
Ideogram just released Ideogram 4.0, a 9.3B-parameter open-weight Diffusion Transformer that outperforms all open-weight image models in LMArena and Design Arena, with a 0.97 X-Omni English OCR text-rendering score (vs. 0.30–0.50).
The engine behind it? AMD Instinct MI350X GPUs, 288 GB of HBM3 memory and 8 TB/s of bandwidth, delivering the throughput enterprise diffusion workloads demand.
Read how @ideogram_ai and AMD are scaling production-grade visual AI from millions to billions of images per month 👉 bit.ly/3TosU1m
1
3
43
AMD and DriveNets are collaborating on a joint reference architecture combining AMD Instinct MI350 series GPUs, AMD ROCm software, RCCL and DriveNets AI Fabric for open, production-scale AI clusters.
Learn more: bit.ly/4wOIzoM
Learn more: bit.ly/4wOIzoM
0
4
45
AMD launches the AMD Helios Rackscale Solution, an open platform connecting 72 AMD Instinct MI455X GPUs for frontier AI inference and training.
Learn more: bit.ly/4wnX0R8
Learn more: bit.ly/4wnX0R8
1
7
71
Overheard at #AdvancingAI
"Helios is the next step in our evolution as a data center leader."
Forrest Norrod shares why AMD Helios is the result of years of planning across compute, networking and software, delivering the foundation for AI at scale.
"Helios is the next step in our evolution as a data center leader."
Forrest Norrod shares why AMD Helios is the result of years of planning across compute, networking and software, delivering the foundation for AI at scale.
0
4
31
AMD and @cerebras Systems are partnering on a new disaggregated AI inference solution.
AMD Helios + Cerebras Wafer-Scale Engine: one integrated workflow delivering up to 5x higher T/s/W. Coming via Cerebras Cloud in H2 2026. #AdvancingAI bit.ly/4x3p38u
AMD Helios + Cerebras Wafer-Scale Engine: one integrated workflow delivering up to 5x higher T/s/W. Coming via Cerebras Cloud in H2 2026. #AdvancingAI bit.ly/4x3p38u
1
4
68
The CPU is central to the agentic AI era.
6th Gen AMD EPYC 9006 Series processors are purpose-built to run AI agents, feed accelerators and power enterprise workloads across cloud, AI and HPC.
Learn more: bit.ly/4yxZaPo
6th Gen AMD EPYC 9006 Series processors are purpose-built to run AI agents, feed accelerators and power enterprise workloads across cloud, AI and HPC.
Learn more: bit.ly/4yxZaPo
1
2
34
Introducing AMD Instinct MI400 Series GPUs.
Built for frontier AI, sovereign AI and HPC, the portfolio combines HBM4 memory, advanced security, scalable system architectures and AMD ROCm software to accelerate AI at every scale.
Learn more: bit.ly/4bYEuq6
Built for frontier AI, sovereign AI and HPC, the portfolio combines HBM4 memory, advanced security, scalable system architectures and AMD ROCm software to accelerate AI at every scale.
Learn more: bit.ly/4bYEuq6
1
4
50
Introducing AMD Helios, a rackscale AI platform built for frontier AI, high-volume inference and foundation-model training.
Built on AMD Instinct GPUs, 6th Gen AMD EPYC CPUs, AMD Pensando networking and AMD ROCm software.
Learn more: bit.ly/4hsfN97
Built on AMD Instinct GPUs, 6th Gen AMD EPYC CPUs, AMD Pensando networking and AMD ROCm software.
Learn more: bit.ly/4hsfN97
0
5
64
Bigger portfolio. Broader story.
Welcome to AMD Data Center.
While our @AMDServer handle stays the same, our new identity reflects a broader focus across the AMD data center portfolio, from AMD EPYC processors and AMD Instinct accelerators to networking and the technologies , and enterprise infrastructure.
Tomorrow at #AdvancingAI 2026, we'll begin sharing the latest innovations shaping the future of the modern data center. Follow along.
Welcome to AMD Data Center.
While our @AMDServer handle stays the same, our new identity reflects a broader focus across the AMD data center portfolio, from AMD EPYC processors and AMD Instinct accelerators to networking and the technologies , and enterprise infrastructure.
Tomorrow at #AdvancingAI 2026, we'll begin sharing the latest innovations shaping the future of the modern data center. Follow along.
0
5
100
The AMD Helios rackscale solution brings together AMD Instinct MI455X GPUs, AMD EPYC CPUs, AMD Pensando networking, and AMD ROCm software to power the next generation of AI.
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
0
4
30
AMD Helios gives partners a foundation to build AI factories with high performance, leading TCO, and open, scalable, serviceable, and secure infrastructure.
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
0
0
13
AMD Helios brings an open ecosystem approach to the rack level. Built on Open Rack Wide architecture, the AMD Helios rackscale solution aligns power, cooling, and serviceability for scalable AI factories.
Tune in to #AdvancingAI on July 23. bit.ly/4yqOh1Z
Tune in to #AdvancingAI on July 23. bit.ly/4yqOh1Z
2
0
10
The AMD Helios rackscale solution is powered by 72 AMD Instinct MI455X GPUs, delivering leadership rack-level performance and massive HBM4 memory for the next era of AI.
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
1
1
17
AI infrastructure is entering a new phase. AMD’s Andrew Dieckmann shares why rack-scale systems are becoming the new building block for AI factories, and why the AMD Helios rackscale solution matters.
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
Tune in July 23 for #AdvancingAI. bit.ly/4yqOh1Z
0
0
8
While everyone is talking about CPUs and agentic AI, AMD is delivering them.
AMD EPYC Server CPUs help power the orchestration, coordination, and execution layer behind agentic AI.
See why CPUs are becoming a critical part of AI infrastructure. bit.ly/4wwEA0c
AMD EPYC Server CPUs help power the orchestration, coordination, and execution layer behind agentic AI.
See why CPUs are becoming a critical part of AI infrastructure. bit.ly/4wwEA0c
0
2
12
Agentic AI requires more than inference.
From orchestration and memory to I/O and concurrency, CPUs play a critical role in helping AI agents operate at scale.
Here are 5 ways CPUs help make agentic AI fast and efficient. bit.ly/4aF6HBN
From orchestration and memory to I/O and concurrency, CPUs play a critical role in helping AI agents operate at scale.
Here are 5 ways CPUs help make agentic AI fast and efficient. bit.ly/4aF6HBN
0
2
9