AmpereOne® processors are the first in a highly evolved family of high-performance, cloud-optimized AI compute products. Beginning with AmpereOne, Ampere has created an AI computing platform that is purpose-built for the most demanding digital services in the world today. Featuring up to 192 Ampere built high-performance cores, AmpereOne® delivers outstanding performance and power efficiency allowing for high server rack density and linearly scalable throughput. These processors are ideal for powering scale out AI enabled web services with AI inference integration, commonly used in today’s machine learning (ML) applications.

Built on Ampere’s unique cloud-native architecture and advanced 5nm process technology, AmpereOne® extends the proven leadership of the Ampere® Altra® family while remaining grounded in three foundational pillars of value:

• Performance: A consistent, high-throughput micro-architecture featuring a pipelined mesh with single threaded cores that ensures predictability under high utilization.
• Efficiency: Exceptional performance-per-watt to optimize energy usage, reduce operational costs, and advance sustainability goals.
• Scalability: Linear scaling delivers same incremental performance for every core added to each parallel workload in elastic containerized or virtualized environments.

With this breakthrough architecture, AmpereOne enables operators and enterprises to dramatically lower the cost of AI compute, meet unprecedented capacity demands, and drive the next generation of AI enabled cloud services such as recommender engines, predictive analytics, natural language processing, and computer vision applications.

AmpereOne® processor architecture features 192 Ampere Arm ISA-compliant cores providing consistent high performance and pipelined cache to every provisioned process or application. Each core has a dedicated low-latency 2 MB L2 cache helping to maintain consistent performance and mitigate noisy neighbor interference.

A combination of high DDR5 memory bandwidth, 4TB of memory capacity, advanced QoS management features, and consistent frequency ensure highly predictable performance for applications in multi-tenant cloud or dense containerized environments.

AmpereOne architecture includes a coherent mesh interconnect with 64 distributed home nodes and directory-based snoop filters enabling seamless, low latency execution for all scheduled processes across the core network.

These unique features of the AmpereOne architecture deliver high throughput for general purpose workloads of all kinds and modern AI enabled services.
Learn more about AmpereOne’s traffic management and QoS techniques.

AmpereOne® pairs high-density compute with class-leading power management features to deliver a cost-effective infrastructure for operators of everything from Cloud data centers to edge-based applications and devices. Uniquely designed power efficient cores allow for high core density reducing rack sprawl, floor space requirements, energy consumption, and cooling requirements driving measurable TCO savings.

AmpereOne system management features dynamically adjust voltage, frequency and thermal margins, keeping every executing core at a consistent frequency and optimal performance/watt, even under peak processor utilization.

For latency sensitive AI tasks like natural language processing, computer vision and recommender engines, efficient power management translates to higher throughput, lower tail latency and capacity to run more models per system without overprovisioning. AmpereOne architecture drives faster, more predictable AI services that scale seamlessly with demand, helping data centers maximize efficiency and reduce infrastructure costs.

See how AmpereOne cuts TCO by 40%

AmpereOne is built for dense containerized deployments. Applications scale linearly as additional single-threaded cores are utilized. This makes AmpereOne an ideal platform for cloud native applications sensitive to latency variability or needing strict SLAs. Services integrated with AI ranging from visual analytics and natural language processing to classical machine learning pipelines are workloads that benefit from this level of consistency, elasticity and determinism.

With 128 lanes of PCIe Gen5, AmpereOne can support up to eight x16 devices attached to a single socket. IT providers have the flexibility to pair AmpereOne with a combination of AI inference accelerators, networking cards and high-performance storage/NVMe drives.

AmpereOne platforms combined with AI accelerators provide operational flexibility to expand model capacity, batch sizes or concurrent sessions on one system. Models can be provisioned according to size and service level agreements without allocating racks to inflexible single purpose GPUs. AmpereOne is the ideal foundation for at-scale AI-enabled web-services and enterprise applications.

* Performance per socket and usage power data is based on estimated SPECrate®2017_int_base (community GCC15) and are subject to change based on system configuration and other factors. Usage Power is defined as average power consumed over time while executing the SPECrate®2017_int_base benchmark.

All data and information contained in or disclosed by this document are for informational purposes only and are subject to change.

This document may contain technical inaccuracies, omissions and typographical errors, and Ampere Computing LLC, and its affiliates (“Ampere”), is under no obligation to update or otherwise correct this information. Ampere makes no representations or warranties of any kind, including express or implied guarantees of noninfringement, merchantability or fitness for a particular purpose, regarding the information contained in this document and assumes no liability of any kind. Ampere is not responsible for any errors or omissions in this information or for the results obtained from the use of this information. All information in this presentation is provided “AS IS”, with no guarantee of completeness, accuracy, or timeliness.

This document is not an offer or a binding commitment by Ampere. Use of the products and services contemplated herein requires the subsequent negotiation and execution of a definitive agreement or is subject to Ampere’s Terms and Conditions for the Sale of Goods.

©2025 Ampere Computing LLC. All Rights Reserved. Ampere, Ampere Computing, AmpereOne®, and the Ampere logo are all registered trademarks or trademarks of Ampere Computing LLC or its affiliates. Arm is a registered trademark of Arm Limited (or its subsidiaries) in the US and/or elsewhere. All other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.