ARM C-series
Mobile processor cores with on-device AI acceleration
Arm Holdings introduced the C-series in 2025 to replace Cortex-A and Cortex-X naming. These 64-bit cores use Armv9.3-A architecture and SME2 technology for faster artificial intelligence tasks on mobile devices.
- Introduced in 2025 to replace Cortex naming
- Uses Armv9.3-A architecture with SME2 support
- Four tiers: Nano, Pro, Premium and Ultra
- Licensed to partners for custom SoC integration
- C1-Ultra succeeds Cortex-X925
What it is
The ARM C-series is a family of processor cores created by Arm Holdings. Launched in 2025, it replaces the older Cortex-A and Cortex-X naming conventions. These chips are built for consumer electronics and mobile phones. They use the Armv9.3-A instruction set and include hardware support for matrix operations. This design helps devices run artificial intelligence tasks locally without needing cloud connections. The series focuses on balancing raw speed with energy efficiency for handheld gadgets.
How it came about
Arm Holdings developed this family to modernise its mobile core lineup. The first group, called C1-series, arrived in September 2025. It marked the debut of the Armv9.3 architecture in commercial designs. The company aimed to simplify its product tiers while boosting AI performance. A second group, the C2-series, followed in September 2026. Both generations rely on the same underlying architectural standards but target different market needs over time.
Inside the chip
All cores in this family are 64-bit processors. They feature Scalable Matrix Extension 2 (SME2) to speed up machine learning workloads. The lineup has four tiers: Nano, Pro, Premium and Ultra. Nano cores use in-order execution for low power. The other three use out-of-order execution for higher speed. Ultra cores offer peak single-threaded performance. Premium cores provide similar speed in a smaller area. Pro cores focus on sustained efficiency for gaming. Nano handles background tasks. A shared interconnect called C1-DSU manages cache and data flow between cores.
How it was made
Arm Holdings does not manufacture these chips itself. Instead, it licenses the hardware designs to other semiconductor companies. These partners integrate the cores into their own system-on-chip products. Manufacturers can choose specific process nodes and customise features. This model allows flexibility for different power targets and production technologies. The C-series is part of the ARM Lumex compute subsystem platform, which offers reference designs for integration.
Where it went
These processors power mobile phones and consumer devices. Ultra cores handle demanding AI jobs and heavy single-threaded tasks. Premium cores fit designs where space is limited but performance must remain high. Pro cores suit applications like video processing and multitasking that need steady power efficiency. Nano cores work in wearables and small gadgets for always-on functions. Chipmakers combine these tiers to create balanced systems for various device classes.
Versions and family
The C1-series launched first with four models. C1-Ultra replaces the Cortex-X925. C1-Pro succeeds the Cortex-A725. C1-Nano takes over from the Cortex-A520. C1-Premium is a new sub-flagship tier. The C2-series arrived later in 2026 using the same architecture. Each variant targets a specific balance of area, power and speed. The Nano tier is unique for using in-order execution, while others use out-of-order designs.
Why it matters
This series shifts how Arm labels its high-performance cores. It integrates AI acceleration directly into the core architecture via SME2. This change supports growing demand for on-device language models and vision tasks. By offering distinct tiers, it gives chip designers more granular control over performance and efficiency. The licensing model continues to enable widespread adoption across the mobile industry.
Questions people ask about ARM C-series
What does the C-series replace?
It succeeds the Cortex-A and Cortex-X naming schemes. Specific models replace older cores like Cortex-X925, Cortex-A725 and Cortex-A520 within the new tier structure.
How does it help with AI?
All cores include Scalable Matrix Extension 2 (SME2). This hardware feature accelerates matrix operations used in large language models, speech recognition and computer vision tasks.
Who manufactures these chips?
Arm Holdings designs the cores but does not make them. Semiconductor companies license the designs and integrate them into their own system-on-chip products using various manufacturing processes.
Good reading and sources
The facts on this page come from the Wikipedia article "ARM C-series", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryChip rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.
It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.