AMD

What AMD does

Advanced Micro Devices (AMD) is a global semiconductor company founded in 1969 and headquartered in Santa Clara, California, that designs high-performance and adaptive computing products spanning four technology pillars: CPUs (EPYC server processors with up to 192 cores on Zen 5, Ryzen desktop and mobile processors with 3D V-Cache, Ryzen Threadripper workstation chips, and Ryzen PRO commercial parts); data center GPU accelerators (AMD Instinct MI300X, MI350/MI355X, and upcoming MI450 series built on CDNA architecture); adaptive SoCs, FPGAs, and DPUs (Versal, Zynq, Kria SOM, Alveo accelerator cards, Pensando Salina DPUs and Pollara 400 NICs); and consumer/professional graphics (Radeon RX 9000/7000 series, Radeon PRO W7000/R9000). The portfolio is unified by the ROCm open-source software platform, AMD Software: Adrenalin Edition with bundled generative AI features, Vitis/Vivado developer toolchains, ZenDNN inference libraries, and the Ryzen AI software stack.

AMD's business model is overwhelmingly hardware-centric, generating revenue through three primary segments: Data Center (EPYC CPUs and Instinct GPUs sold to hyperscalers and large enterprises), Client and Gaming (Ryzen CPUs, Radeon GPUs, and gaming handheld processors sold via OEMs and retailers), and Embedded/Adaptive Computing (Versal SoCs, FPGAs, Kria SOMs, Pensando DPUs, and Solarflare adapters serving aerospace, automotive, broadcast, industrial, and telecom verticals). Go-to-market combines direct enterprise sales to hyperscalers (e.g., OpenAI, Meta, Oracle, Rackspace, JPMorganChase), deep OEM partnerships embedding AMD silicon into Dell PowerEdge, HPE Cray, Supermicro, Lenovo, HP, ASUS, Valve, and MiTAC platforms, a global channel network of distributors and resellers spanning 50+ offices across 31 countries, and developer/community engagement via ROCm, AMD Developer Central, and the AI Developer Program. AMD reports Q1 FY2026 revenue of $10.25B (up 38% YoY) with Data Center revenue of $5.78B (up 57% YoY) and non-GAAP EPS of $1.37 (up 43% YoY), reflecting a strategic transition from chip supplier to AI infrastructure platform provider anchored by multi-gigawatt hyperscaler commitments.

Name

AMD

Legal name

Advanced Micro Devices, Inc.

Website

https://amd.com

Company type

Public

Founded year

1969

Operating status

Operating

Headcount range

5,001–10,000 employees

Short description

Advanced Micro Devices (AMD) is a global semiconductor company designing CPUs (EPYC, Ryzen), GPUs (Instinct, Radeon), and adaptive SoCs for hyperscalers, OEMs, and consumer markets, with Q1 FY2026 revenue of $10.25B driven by data center and AI infrastructure contracts.

Ownership category

akta.pro rank

Product category

Semiconductor Design and Manufacturing

NAICS

Computer Storage Device Manufacturing (334112), Computer and Peripheral Equipment Manufacturing (33411)

akta.pro primary industry

Compute & Acceleration ICs (CPU/GPU/AI/FPGA/ASIC) (HDAHABAC)

akta.pro secondary industry

System-on-Chip (SoC) Platforms (Mobile/Consumer/Embedded) (HDAHAJAC)

Keywords

• Data center processors
• AI GPU accelerators
• Adaptive SoCs FPGAs
• Client gaming CPUs
• Open GPU software

Headquarters

HQ city

Santa Clara

HQ country

United States

HQ region

North America

Offices2 records

Markets served

GTM type

B2B and B2C

Offering type

Hardware or Manufacturing

Cost components

Technology or R&D, Supply Chain, Personnel, Marketing or Sales, Operations, Infrastructure

Revenue model

1. Data Center Hardware Sales: Sales of EPYC server CPUs and Instinct GPUs to hyperscalers and enterprises. Q1 FY2026 Data Center revenue reached $5.78B (up 57% YoY), representing 56% of total revenue, driven by EPYC processor adoption and large GPU deals including OpenAI, Meta, Oracle, and Rackspace deployments.
2. Client and Gaming Hardware Sales: Sales of Ryzen desktop/mobile CPUs, Radeon graphics cards, and gaming handheld processors to consumer and commercial PC OEMs and retailers. AMD's total revenue reached $10.3B in Q1 2026 (up 38% YoY).
3. Embedded and Adaptive Computing Sales: Sales of Versal adaptive SoCs, FPGAs, Kria SOMs, Pensando DPUs and Solarflare Ethernet adapters serving aerospace, automotive, broadcast, industrial, telecom, and test/measurement verticals via partners like Mobilint and DFI.

Pricing tiers

Go-to-market motion4 records

Core offering

AMD designs and sells high-performance and adaptive semiconductor products including server CPUs (EPYC), client and gaming CPUs (Ryzen, Threadripper, Ryzen PRO), data center GPU accelerators (Instinct), adaptive SoCs and FPGAs (Versal, Zynq, Kria), DPUs and Ethernet adapters (Pensando, Solarflare), and consumer/professional graphics (Radeon RX, Radeon PRO). The portfolio is unified by the open-source ROCm software platform, AMD Software: Adrenalin Edition graphics drivers, Vitis and Vivado developer toolchains, FidelityFX FSR upscaling technology, and architectures including Zen 5, RDNA, and CDNA.

Product overview

AMD operates a modular, platform-plus-products portfolio spanning high-performance and adaptive computing. The core silicon offering divides into four pillars: (1) CPUs (EPYC server processors including 9005/9004/8005/8004/4005/4004 series with Zen 5 architecture; Ryzen consumer desktop processors 9000/8000/7000 with 3D V-Cache on Ryzen 9 9950X3D2 and Ryzen 7 9850X3D; Ryzen AI laptop processors with integrated NPUs; Ryzen PRO commercial processors including AI PRO 400/300/8040/7040 series; Ryzen Threadripper and Threadripper PRO 9000/7000/5000 WX workstation processors; and EPYC/Ryzen Embedded); (2) Data center GPU accelerators (AMD Instinct MI300X, MI350/MI350P/MI355X built on CDNA architecture, and upcoming MI450/Helios rack-scale AI platform with 6 gigawatts committed by OpenAI and Meta); (3) Adaptive compute (Versal adaptive SoCs, FPGAs, Kria SOM, AMD Alveo X3 accelerator cards, AMD Pensando Salina DPU and Pollara 400 NIC, Solarflare Ethernet adapters); and (4) Graphics (Radeon RX 9000/7000 consumer cards on RDNA 4/RDNA 3, and Radeon PRO AI PRO R9000 plus W7000/W6000 workstation cards). The portfolio is unified by the AMD ROCm open software platform, AMD Software: Adrenalin Edition graphics driver with bundled generative AI tools, Vitis/Vivado developer toolchains for adaptive compute, Ryzen AI Software for on-device ML, ZenDNN inference library, Zen Software Studio for server optimization, AMD Pensando SSDK for DPU development, and FidelityFX FSR upscaling technology. AMD has expanded through acquisitions of MK1 (AI inference), MEXT (memory tiering), Silo AI, and ZT Systems, while partner ecosystems span Dell PowerEdge, HPE Cray GX5000 with EPYC Venice, Rackspace Enterprise AI Cloud, Valve Steam Machine, JWIPC workstations, and Ultra Ethernet Consortium founding membership.

Differentiator

Problem solved

Functional benefit

Brands

• AMD EPYC: Server CPU family for data centers, cloud, and AI workloads (series include 9005, 9004, 8005, 8004, 4005, 4004)
• AMD Ryzen
• AMD Ryzen Threadripper
• AMD Ryzen PRO
• AMD Instinct
• AMD Radeon
• AMD Radeon PRO
• AMD Alveo
• AMD Versal
• AMD Kria
• AMD ROCm
• AMD FidelityFX Super Resolution (FSR)
• AMD FreeSync
• AMD Software: Adrenalin Edition
• AMD Ryzen Master
• AMD Vitis
• AMD Vivado
• AMD HYPR-RX
• AMD Silo AI
• AMD Radeon ProRender

Products and services

• AMD EPYC Server CPUs High-performance x86 server processors designed for data center, cloud, virtualization, AI, and enterprise workloads. Portfolio includes EPYC 9005 (Zen 5), 9004, 8005, 8004, 4005, and 4004 series, with up to 192 cores. For hyperscalers, cloud providers, and enterprise data centers.
• AMD Ryzen Desktop Processors Consumer and enthusiast desktop processors including Ryzen 9000 series (Zen 5), Ryzen 8000 series, Ryzen 7000 series, and the high-end Ryzen 9 9950X3D2 and Ryzen 7 9850X3D featuring 3D V-Cache technology for gaming performance. For enthusiast PC builders and gamers.
• AMD Ryzen Laptop Processors Mobile processors including Ryzen AI Max 300 series, Ryzen AI 400 series, Ryzen AI 300 series, and Ryzen 200 series with integrated NPUs for on-device AI acceleration in laptops from OEMs. For OEM laptop manufacturers and consumer mobile users.
• AMD Ryzen AI Processors AI-enabled client processors featuring dedicated NPUs (e.g., delivering up to 80 TOPS in ASUS ROG laptops) for on-device generative AI, Copilot+ PC experiences, and AI PC workloads.
• AMD Ryzen PRO Processors Commercial-grade processors for business laptops, desktops, and workstations, including Ryzen PRO 9000 series, Ryzen AI PRO 400 series, Ryzen PRO 8040, Ryzen PRO 7040, and Ryzen AI PRO 300 series, with PRO Technologies for enterprise manageability and security. For commercial PC buyers and IT departments.
• AMD Ryzen Threadripper Processors High-core-count processors for professional workstations and compute-intensive workloads including rendering, simulation, engineering, and content creation. Portfolio spans Threadripper 9000 series, Threadripper PRO 9000 WX series, Threadripper PRO 7000 WX, Threadripper PRO 5000 WX, and Threadripper 7000 series. For professional workstation users and content creators.
• AMD Instinct GPUs Data center GPU accelerators built on AMD CDNA architecture (CDNA 3 in MI200 series, CDNA 4 referenced), with Matrix Core technology and full precision support for AI training, inference, and HPC workloads. MI300X referenced as platform for third-party inference engines; MI350/MI350X/MI355X series offer up to 144GB HBM3e and 4,600 peak teraflops. For hyperscalers, AI labs, and HPC customers.
• AMD Alveo Data Center Accelerator Cards FPGA-based adaptive accelerator cards for data center workloads including networking, storage, and compute acceleration. Alveo X3 series (e.g., A-X3522) is offered alongside Alveo App Store for pre-built applications deployable on-premises and in cloud.
• AMD Pensando DPUs and Network Adapters AMD Pensando Salina DPU and Pensando Pollara 400 network interface cards are fully programmable P4-based data processing units optimized to deliver cloud, compute, network, storage, and security services. Solarflare Ethernet adapters with Onload software accelerate network-intensive applications.
• AMD Radeon RX Graphics Cards Consumer gaming GPUs based on RDNA architecture (RDNA 3 in RX 7000 series, RDNA 4 in RX 9000 series including RX 9070, RX 9070 XT, RX 9070 GRE, RX 9060 XT, RX 9060, and RX 9060 XT LP) with AI-accelerated FSR upscaling and HYPR-RX technologies. For gamers and PC enthusiasts.
• AMD Radeon PRO Graphics Cards Professional workstation GPUs including Radeon AI PRO R9000 series (R9700, R9700S, R9600), Radeon PRO W7000 series (W7800, W7900), and Radeon PRO W6000 series, certified for ISV applications and optimized for demanding creative, engineering, and visualization workflows.
• AMD Versal Adaptive SoCs Adaptive SoCs featuring AI Engines, design security, DSP, functional safety, high-speed serial, memory, and power-efficiency technologies. Includes Gen2 AI Edge Series targeting edge inference workloads. For aerospace, automotive, broadcast, industrial, and embedded customers.
• AMD Kria SOM System-on-Module (SOM) products for edge AI and embedded applications, including Kria K26 series with KV260 Vision AI and KR260 Robotics starter kits.
• AMD ROCm Open Software Platform AMD's open-source GPU computing software ecosystem built on a code-once, use-everywhere approach for HPC and AI workloads, including the AMD ROCm AI Developer Hub, Infinity Hub software containers, and Enterprise AI Suite for production AI deployment. ROCm 7 referenced as latest version. For AI developers and HPC users.
• AMD Software: Adrenalin Edition Desktop GPU control software for optimizing gaming, display configuration, driver updates, and optional generative AI tools that work locally on the device. Includes HYPR-RX technology and is bundled with AI features. For Radeon GPU owners.
• AMD Vitis and Vivado Software Platforms Vitis unified software platform and Vivado ML-based design suite for software developers and hardware engineers building accelerated applications targeting AMD Alveo adaptive accelerators, SoCs, FPGAs, and SOMs. Vitis AI provides integrated AI inference development environment with optimized IP, tools, libraries, and models. For adaptive computing developers.
• AMD Ryzen Master Utility Overclocking and monitoring utility for AMD CPUs providing CPU overclocking, parameter monitoring, memory overclocking, and the AMD Ryzen Master Monitoring SDK for adding processor/memory functions to third-party utilities. For enthusiast PC users.
• AMD Ryzen AI Software Software toolchain that takes ML models trained in PyTorch or TensorFlow and runs them on select laptops powered by AMD Ryzen AI technology, enabling on-device AI inference. For AI PC developers.
• AMD Ryzen Z Series Handheld Processors Processors including Ryzen Z2 Extreme powering handheld gaming devices like Lenovo Legion Go 2 and SteamOS-based handhelds with RDNA 3 graphics.
• AMD Embedded Processors (EPYC and Ryzen) x86 embedded microprocessors (including OPNs with S suffix) and embedded graphics/media processing for industrial, aerospace and defense, automotive, telecom, healthcare, and other vertical applications.
• AMD Sovereign AI Solutions Sovereign AI solutions positioning AMD portfolio of processors, accelerators, and networking as foundational classical infrastructure for regulated and national AI deployments.
• AMD Silo AI AI lab and consulting capability integrated into AMD offering custom AI models and solutions for enterprise customers.
• AMD Helios Rack-Scale AI Platform Rack-scale AI platform supporting 6 gigawatts of AMD-powered infrastructure committed by OpenAI and Meta over multiple years. Anchors Oracle's 50,000-GPU Helios supercluster deployment and underpins AMD's transition from semiconductor supplier to AI infrastructure company.
• AMD FidelityFX Super Resolution (FSR) Suite of visual effects including FidelityFX Super Resolution (FSR) ML-based upscaling technology (FSR 4.1 with INT8 model for RDNA 3 and FP8 for RDNA 4) and FSR Redstone technology for advanced upscaling and frame generation.

Quantifiable outcome

• Data Center revenue grew 57% YoY to $5.78B in Q1 FY2026
• +6 more outcomes

Named customers16 records

Ideal customer profiles6 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration14 records

AI capability12 records

Feature11 records

Partnerships

31 partnerships are on record, tiered flagship, core and minor.

• OpenAIflagshipStrategic or Co-development PartnerStrategic AI infrastructure partnership covering 6 gigawatts of AMD Instinct GPU deployment across multiple years. Anchors AMD's Helios rack-scale AI platform and represents a major transition from chip supplier to AI infrastructure provider for AMD. The partnership spans deployment of AMD MI450 accelerators and is one of the largest GPU deals announced in the industry.
• Meta PlatformsflagshipStrategic or Co-development PartnerMulti-year commitment for up to 6 gigawatts of AMD Instinct GPU deployment for Meta's AI infrastructure. The deal anchors AMD's loaded customer pipeline and signals Meta's multi-vendor GPU sourcing strategy to reduce dependence on Nvidia.
• OracleflagshipStrategic or Co-development PartnerOracle built a 50,000-GPU Helios supercluster using AMD MI450 accelerators. The deployment represents a flagship AI infrastructure partnership combining AMD's rack-scale platform with Oracle's cloud capabilities.
• Dell TechnologiesflagshipOEM/ Whitelabel/ Licensing PartnerOEM partnership integrating AMD Instinct MI350P PCIe GPUs into Dell PowerEdge servers starting July 2026. The collaboration offers up to 4,600 peak teraflops and 144GB HBM3e memory for on-premises generative and agentic AI deployments. Dell also enhances its AI Platform with AMD to support pilot-to-production workloads.
• Rackspace TechnologycoreStrategic or Co-development PartnerDefinitive agreement for phased deployment of 30 MW AMD-based AI compute capacity across Rackspace's global data centers from late 2026 through 2028. Integrates AMD Instinct GPUs and EPYC CPUs within Rackspace's Enterprise AI Cloud architecture for sovereign and regulated enterprise AI deployment.
• HPE (Hewlett Packard Enterprise)flagshipOEM/ Whitelabel/ Licensing PartnerHPE Cray GX5000 supercomputing platform at HPE Discover 2026 features AMD EPYC Venice processors delivering up to 81,920 CPU cores in a single 42U rack with 1.28PB liquid-cooled RAM. HPE is also leveraging Juniper Networks acquisition for autonomous networking and AI infrastructure.
• LenovocoreOEM/ Whitelabel/ Licensing PartnerLenovo ThinkStations and ThinkPads powered by AMD Threadripper and Ryzen processors; Lenovo WenTian WA7785a G3 server achieved 8,000 tokens/s throughput running DeepSeek 671B parameter model on AMD Instinct GPUs.
• SupermicrocoreOEM/ Whitelabel/ Licensing PartnerOEM partnership for AMD EPYC servers; Supermicro MicroCloud Servers Advanced by AMD EPYC CPUs is featured in AMD webinars and Supermicro is a data center partner ecosystem member.
• HP Inc.coreOEM/ Whitelabel/ Licensing PartnerOEM partnership for AMD Ryzen, Ryzen PRO, and Ryzen AI processors across HP commercial PCs and workstations; joint webinars and HP-AMD commercial PC portfolio collaborations.
• ASUScoreOEM/ Whitelabel/ Licensing PartnerOEM partnership for AMD Ryzen, Ryzen Threadripper, and Radeon products including ROG gaming laptops and ZenBook/ProArt creator laptops. ASUS was recognized as AMD partner in PCIe and gaming collaborations.
• ValvecoreOEM/ Whitelabel/ Licensing PartnerValve Steam Machine living-room gaming console features AMD's semi-custom Zen 4 CPU and RDNA 3 GPU, priced at $1,049 (512GB) and $1,349 (2TB) launching June 29-30, 2026. Valve is collaborating with AMD to bring improved FSR 4 upscaling features to the system, including RDNA 3.5 APU support.
• MiTAC Computing TechnologycoreTechnology or IntegrationStrategic collaboration with AMD and Intel on Agentic AI-ready infrastructure showcased at ISC 2026. Products include G4826Z5 liquid-cooled server with AMD Instinct MI355X GPUs and G4520G6 with Intel Xeon 6, designed for HPC and AI workloads with up to 6TB DDR5 memory and PCIe Gen5.
• Advanced Semiconductor Engineering (ASE)coreTechnology or IntegrationAMD committed over $10 billion to expand Taiwan's semiconductor and AI ecosystem, collaborating with ASE and SPIL on advanced chip packaging technologies including Elevated Fanout Bridge (EFB) for next-generation AI systems.
• Samsung ElectronicscoreTechnology or IntegrationSamsung pursuing long-term HBM4 and HBM4E supply agreements with AMD (alongside Nvidia, Broadcom, Google). Samsung began mass-producing HBM4 chips in February 2026, becoming the first in the industry to do so. AMD EPYC Venice processors incorporate Samsung SSDs in HPE Cray GX5000.
• SK HynixcoreTechnology or IntegrationSK Hynix is a primary HBM supplier for AMD's AI accelerator GPUs. SK Hynix slowing HBM4 expansion in June 2026 contributed to AMD stock volatility amid concerns about AI memory demand.
• TSMCflagshipTechnology or IntegrationTSMC manufactures chips for AMD (alongside Apple, Nvidia, Amazon, Google, Microsoft). TSMC Q1 fiscal 2026 revenue of $35.9B (up 41% YoY) reflects strong AI demand for advanced N2 node technology.
• IntelflagshipStrategic or Co-development PartnerAMD and Intel jointly introduced AI Compute Extensions (ACE), a new x86 instruction set specification to accelerate AI workloads on CPUs. AMD took primary authorship (8 of 11 authors) on the ACE v1.15 specification, adding 8 new tile registers with 16x compute density improvement. Joint effort to defend x86 architecture against Arm in AI computing markets.
• JPMorganChasecoreStrategic or Co-development PartnerCollaboration on quantum-HPC-AI integration as part of AMD's hybrid quantum-classical computing architecture advocacy. Establishes AMD as the classical infrastructure layer supporting quantum systems.
• Oak Ridge National LaboratorycoreStrategic or Co-development PartnerCollaboration on quantum-HPC-AI integration research and AMD HIP software adoption for rapid migration of scientific workloads; ORNL exascale deployment uses AMD technology.
• IBMcoreStrategic or Co-development PartnerCollaboration on quantum-HPC-AI integration as part of AMD's hybrid quantum-classical computing strategy.
• Ultra Ethernet ConsortiumcoreTechnology or IntegrationAMD is a founding member of the Ultra Ethernet Consortium alongside Arista, Broadcom, Cisco, HPE, Intel, Meta, and Microsoft; released formal Specification 1.0 on June 11, 2025 to address networking bottlenecks in AI data centers.
• JWIPCminorOEM/ Whitelabel/ Licensing PartnerJWIPC showcased OpenClaw local AI solution lineup across Intel, AMD, and NVIDIA platforms at COMPUTEX 2026, supporting local AI infrastructure and sovereign AI deployment.
• GigabyteminorOEM/ Whitelabel/ Licensing PartnerOEM partner for AMD Ryzen X3D processors; Gigabyte X3D Turbo Mode 2.0 platform-level performance feature enhances AMD Ryzen X3D processor capabilities with up to 34% gaming performance improvement.
• Amazon Web Services (AWS)flagshipChannel Partner/ Reseller/ DistributorCloud partner offering AMD EPYC instances; AMD Arena training course on "Convert to AMD Instances in Amazon EC2" supports customer migration to AMD-based cloud infrastructure.
• Microsoft AzureflagshipChannel Partner/ Reseller/ DistributorCloud partner offering AMD EPYC instances for enterprise customers; Microsoft is among data center ecosystem partners.
• Google CloudflagshipChannel Partner/ Reseller/ DistributorCloud partner offering AMD EPYC instances for enterprise customers; AMD Arena training and Google Cloud integration supports EPYC-based cloud deployments.
• MEXTcoreStrategic or Co-development PartnerAMD acquired MEXT on June 15, 2026. MEXT is an AI-driven memory optimization technology startup whose predictive data prefetching platform makes flash storage behave more like DRAM, improving utilization by 20-50% and addressing HBM memory bottlenecks.
• MK1coreStrategic or Co-development PartnerAMD acquired MK1, a California-based (Mountain View) AI startup specializing in AI inference and reasoning software. MK1's software currently handles over 1 trillion tokens daily on AMD Instinct GPUs, expanding AMD's AI software portfolio.
• AT&S Austria Technologie & SystemtechnikcoreTechnology or IntegrationMulti-year supply commitments from AMD for AT&S Kulim manufacturing facility expansion, driven by AI and data center infrastructure demand. AT&S stock surged 82.72% over 30 days following the AMD announcement.
• CiscocoreOEM/ Whitelabel/ Licensing PartnerOEM/data center ecosystem partner for AMD EPYC solutions; Cisco is featured in AMD training courses and ecosystem partner listings.
• DFI (Industrial Edge AI)minorOEM/ Whitelabel/ Licensing PartnerDFI showcased Industrial Edge AI portfolio at Automate 2026 in collaboration with South Korean AI chip provider Mobilint using AMD processors for smart manufacturing applications.

Scale indicators12 records

Recent moves7 records

Expansion highlights6 records

Direct peers

• NVIDIA: Direct competitor across data center GPUs (Instinct vs. H100/B200) and AI accelerators, both targeting the same hyperscaler training and inference workloads with rival software platforms (ROCm vs. CUDA).
• Intel: Direct competitor in server CPUs (Xeon vs. EPYC) and client PCs, now also a joint ACE specification partner; AMD's server share gains are coming at Intel's expense in x86 data center.

Broad incumbents

• Broadcom: Competes in custom AI silicon (ASICs) for hyperscalers and in networking silicon for AI clusters, offering an alternative path to GPU-based AI infrastructure alongside AMD's Instinct and Pensando offerings.
• Apple: Designs custom Arm-based silicon (M-series, A-series) that competes with AMD Ryzen in premium laptops and desktops; not a direct data center competitor but materially shapes the high-end client compute market.

Emerging players

• Qualcomm: Competes with AMD Ryzen AI in AI-enabled mobile and edge NPUs for Copilot+ PCs and on-device AI inference, with growing ambitions in data center CPUs (SAECP-AI-100).
• Marvell Technology: Competes in custom AI silicon for hyperscalers and in DPUs/networking, intersecting with AMD Pensando's data processing unit and custom compute offerings.
• Arm Holdings: Architectural competitor challenging x86 in data center and client CPUs (NVIDIA Grace, Microsoft Cobalt, AWS Graviton); the ACE specification is AMD and Intel's coordinated defense against Arm in AI workloads.

Regional players

• Samsung Electronics: HBM memory supplier to AMD and a regional competitor in memory and foundry services; also producing consumer SSDs and competing in adjacent semiconductor categories serving AI infrastructure.

Others

• Texas Instruments: Competes with AMD Embedded (EPYC/Ryzen Embedded) in industrial, automotive, and aerospace markets, with overlapping reach into analog and embedded processing for edge applications.
• Micron Technology: Key HBM memory supplier for AMD Instinct AI accelerators and adjacent provider of high-bandwidth memory; comparable as an enabler of AI infrastructure and direct beneficiary of the same GPU ramp.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat7 records

Key risks6 records

Key highlights7 records

Customer concentration

• Xx.com/amd
• YouTubeyoutube.com/user/amd
• Facebookfacebook.com/AMD
• LinkedInlinkedin.com/company/amd
• Instagraminstagram.com/amd

Revenue estimate

Valuation estimate

Number of profiles

Profiles31 records

Subsidiaries6 records

Funding overview

Funding rounds4 records

Investors3 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A19 records

Investments36 records

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about AMD