ARM Holdings

ARM Holdings designs Central Processing Unit (CPU) architectures and related IP for system-on-chip (SoC) and semiconductor vendors across mobile, infrastructure, automotive, embedded, and emerging Artificial Intelligence (AI) workloads.

• Licensable CPU, Graphics Processing Unit (GPU), and Neural Processing Unit (NPU) architectures and cores for SoC designers (compute IP)
• Platform specifications and system architectures for mobile, client, infrastructure, and automotive silicon
• Development tools, compilers, and software frameworks for building and optimizing applications on Arm-based processors (developer tooling)
• Ecosystem enablement with reference designs, security frameworks, and compatibility standards for Arm-based hardware and software
• Support for partners across smartphones, data centers, edge computing, Internet of Things (IoT), and automotive electronic control units

More About ARM Holdings

ARM Holdings provides processor and system IP that semiconductor companies and device manufacturers license and integrate into custom system-on-chip designs. In enterprise and institutional environments, Arm-based processors are present in servers, storage controllers, networking hardware, edge gateways, and a range of embedded and automotive systems. Customers use Arm CPU cores and related IP as building blocks within their own silicon, enabling consistent instruction set architecture across devices in data center, edge, and client deployments.

Arm defines and maintains the Arm architecture (processor instruction set architecture category), which specifies the instruction set, privilege levels, memory model, and system behavior that Arm-compatible CPUs implement. On top of this architecture, Arm offers CPU core IP for general-purpose compute, GPU IP for graphics and heterogeneous compute (graphics and compute IP category), and NPU or Machine Learning (ML) accelerator IP for ML workloads (AI accelerator IP category). These IP blocks are licensed by chip vendors that integrate them into SoCs serving markets such as smartphones, infrastructure, automotive domain controllers, and industrial control.

For enterprise workloads, Arm architecture is used in server-class processors that run Linux, containerized platforms, cloud-native applications, and common enterprise software stacks. The architecture supports standard server features such as virtualization, multi-core scaling, and 64-bit addressing, enabling deployment in data center and cloud environments alongside x86-based systems. Many operating systems, hypervisors, and language runtimes provide support for Arm, which allows enterprises to treat Arm-based servers as part of heterogeneous compute fleets.

ARM Holdings provides reference system architectures and platform specifications that guide how Arm CPUs, memory subsystems, interrupt controllers, and peripheral buses are combined in a coherent system (system architecture category). The company also defines security models such as hardware-enforced execution realms and trusted execution environments (security architecture category), which system designers can adopt to separate secure workloads from general-purpose code. These architectural specifications enable consistent software support and facilitate firmware, Operating System (OS), and hypervisor development across multiple SoC implementations.

Developer tooling from Arm includes compilers, performance analysis utilities, debuggers, and software libraries (developer tools category), which help software teams optimize workloads for Arm-based processors in embedded, real-time, and high-level OS environments. Alongside this, Arm works with ecosystem partners to provide reference boards, software development kits, and compatibility test programs that validate hardware and software interoperability for Arm-based solutions.

In a technology directory or marketplace, ARM Holdings fits into categories such as semiconductor IP (processor and system IP), compute infrastructure (Arm-based server and edge compute architectures), embedded and IoT platforms (microcontroller and real-time architectures), automotive compute platforms (domain and zonal controller architectures), and AI infrastructure (AI accelerator and NPU IP). Enterprises and OEMs engage with Arm at the architecture and IP level, using its designs as the basis for custom silicon that runs workloads from mobile applications to data center services and safety-related automotive software.