Redefining Mobile Computing with Arm v9.3 Architecture
The Arm C1 CPUs ditch the traditional Cortex-X and A series branding in favor of a fresh naming convention that includes the entire range: Ultra, Premium, Pro, and Nano cores. Each variant is optimized to work harmoniously within clusters, enabling device manufacturers to tailor compute performance for specific workloads and power envelopes. Built on the latest Arm v9.3 architecture, the C1 cores offer next-level instruction-per-cycle (IPC) improvements and integrate cutting-edge AI acceleration technologies such as Arm’s Scalable Matrix Extension 2 (SME2).
Performance Across the C1 Family
- C1-Ultra: The flagship core, boasting up to 25% higher single-threaded performance compared to the Cortex-X2 from the previous generation. Ideal for intensive tasks such as high-frame-rate gaming, complex AI models, and video editing.
- C1-Premium: A mid-tier core focusing on balanced power and performance, offering around 16% uplift in gaming and accelerated AI processing.
- C1-Pro: Designed for steady performance in thermally constrained environments, optimized for common mobile workloads with AI improvements.
- C1-Nano: Specialized for energy-efficient operation, perfect for background and always-on tasks with up to 26% better power efficiency over the Cortex-A520.
Each core type supports scalable cache architectures, including up to 3MB L2 size with shared L3, and features advanced instruction prefetching and branch prediction for maximum efficiency.
Massive Leap in AI Computation
With the integration of Arm’s SME2, the C1 series is engineered for low-latency AI inference right within the CPU cluster. This capability means that smartphones and edge devices can handle increasingly sophisticated AI workloads such as speech recognition, real-time translation, and image processing without offloading tasks to remote servers, thus enhancing privacy and reducing lag.
Arm claims the C1 cluster can deliver up to 5 times the AI performance of previous designs. Supported by a new compute subsystem platform called Lumex CSS, the CPUs pair with the new Mali G1 GPU series—which themselves offer up to 40% better ray tracing performance—for a comprehensive AI and graphics solution.
Energy Efficiency and Sustainable Innovation
In an era where mobile device power consumption is critical, Arm has prioritized efficiency in the C1 series. The C1-Ultra achieves a 25% uplift in single-core performance per watt compared to its predecessor, allowing smartphones to maintain high performance while extending battery life. The smaller C1-Premium and C1-Nano cores focus heavily on reducing power draw during less demanding activities, including always-on AI and background processes.
Ecosystem and Industry Adoption
Arm’s architecture licensing model means that multiple chipset manufacturers such as Qualcomm, MediaTek, and Samsung are expected to adopt the C1 CPUs and Mali G1 GPUs in their upcoming processors. The improved AI-ready platform is also attracting software partners, with support for SME2 being integrated into popular AI frameworks by Google, Microsoft, Alibaba, and Meta, ensuring developers can leverage these capabilities out of the box.
Arm’s C1 Series Paves the Way for Smarter, More Efficient Mobile AI
Arm’s next-generation C1 CPU family sets a new benchmark for mobile processing in 2025 and beyond. With significant improvements in AI inference speed, energy efficiency, and performance scaling, the C1 series offers a versatile foundation for future devices. Combined with the Mali G1 GPUs and Lumex CSS platform, Arm is positioning itself at the forefront of mobile AI innovation, empowering manufacturers to deliver rich, responsive experiences while keeping battery life in check.
The adoption of the C1 CPUs and accompanying GPU lineup is expected to drive forward the evolution of AI-powered smartphones, tablets, and other edge devices, heralding a new era of on-device intelligence capable of accelerated, efficient computing at the fingertips of consumers worldwide.
