Ampere Processing Platforms
Ampere Processing Platforms
Accelerating AI and redefining efficient computing
Accelerating AI and redefining efficient computing
Performance
Built for performance. Ampere’s single-threaded, highly parallel cores are designed to deliver consistently fast, low-latency processing from cloud to edge. Designed for AI Compute means that you get high inference throughput with more agents per server – without the jitter, contention or latency spikes of legacy architectures.
Efficiency
Get more done with less. Ampere’s efficient architecture uses significantly less power and resources while delivering better performance, helping lower energy costs and reducing carbon footprint. Designed for AI Compute means more inference tokens per server, driving down CapEx and OpEx while maximizing how many AI models you can deploy per rack.
Scalability
Grow seamlessly from cloud to edge. Ampere’s elastic architecture scales linearly while maintaining optimized performance and high compute density. Defined for AI Compute means no more reliance on rigid, single-use GPU infrastructure. Instead, you get simple provisioning that adapts to evolving workloads with maximum flexibility.
AmpereOne® Platforms
AmpereOne® Platforms
AmpereOne family includes two server class platforms: AmpereOne and AmpereOne M. AmpereOne platforms offer the best compute efficiency, container/VM density for both traditional workloads and legacy AI models. AmpereOne M platforms deliver advanced compute for dense AI environments deploying multi-modal and agentic AI based LLM services with the lowest cost per inference session available.
AmpereOne family includes two server class platforms: AmpereOne and AmpereOne M. AmpereOne platforms offer the best compute efficiency, container/VM density for both traditional workloads and legacy AI models. AmpereOne M platforms deliver advanced compute for dense AI environments deploying multi-modal and agentic AI based LLM services with the lowest cost per inference session available.
AmpereOne®
Most versatile and efficient for cloud & AI workloads
- 96 - 192 Cores
- 2MB Private L2 Cache per Core
- 64MB System Level Cache
- 8 channel DDR5 – up to 4TB
- 128 lanes PCIe Gen5
- 200 - 400W TDP
AmpereOne® M
Optimized for volume AI inference workloads
- 96 - 192 Cores
- 2MB Private L2 Cache per Core
- 64MB System Level Cache
- 12 channel DDR5 – up to 3TB
- 96 lanes PCIe Gen5
- 250 - 425W TDP
Future of AI Compute
AmpereOne® Aurora
Future of AI Compute
AmpereOne® Aurora
Ampere designed cores, mesh, and chiplet interconnect combined with our innovative AI acceleration equals a revolutionary AI Compute solution.
Ampere designed cores, mesh, and chiplet interconnect combined with our innovative AI acceleration equals a revolutionary AI Compute solution.
Ampere Altra® Platforms
Ampere Altra® Platforms
Ampere Altra platforms deliver extremely efficient, high-performance computing especially for power constrained racks and other sensitive environments. Widely adopted in Telecom, Networking, Autonomous Vehicles and Edge AI applications, Ampere Altra offers scalable & efficient performance for traditional and AI enhanced services.
Ampere Altra platforms deliver extremely efficient, high-performance computing especially for power constrained racks and other sensitive environments. Widely adopted in Telecom, Networking, Autonomous Vehicles and Edge AI applications, Ampere Altra offers scalable & efficient performance for traditional and AI enhanced services.
Ampere Altra®
Low power, high density AI Compute from Cloud to Edge
- 32 - 128 Cores
- 1MB Private L2 Cache per Core
- 16-32MB System Level Cache
- 8 channel DDR4 - up to 4TB
- 128 lanes PCIe Gen4
- 45W - 250W TDP
Ampere Altra® is revolutionizing edge services
- More cores for enhanced AI services
- Industry leading power efficiency
- Server class predictable processing
- Flexibility for many attached IO devices
Ampere Computing LLC
4655 Great America Parkway Suite 601
Santa Clara, CA 95054
