Ampere® Altra® AI Inference Servers
Supermicro offers customers access to Ampere® Altra® family of processors and AmpereOne® family of processors in its MegaDC server series supporting configurations from 32 to 192 cores per CPU socket.
AI Infrastructure Hardware
AI accelerator for genAI – ARA-2
Meet the Kinara Ara-2 AI processor, the leader in edge AI acceleration. This 40 TOPS powerhouse tackles the massive compute demands of Generative AI and transformer-based models with unmatched cost-effectiveness.
AI accelerator for computer vision – ARA-1
Kinara Ara-1 edge AI processors are the engines powering the next generation of smart edge devices. Built around a flexible and efficient dataflow architecture and supported by a comprehensive SDK, Ara-1 processors deliver the performance and responsiveness needed for real-time AI computing and decision-making.
Ampere AI Inference Servers
The new Ampere servers configured by ASA Computers feature Cloud Native Processors that offers industry-leading core density, server efficiency, and per-rack performance, with up to 192 cores providing the best performance/$ compute for AI inferencing.
Ampere Optimized PyTorch
Ampere’s inference acceleration engine is fully integrated with Pytorch framework. PyTorch models and software written with PyTorch API can run as-is, without any modifications.
Ampere Optimized TensorFlow
Ampere’s inference acceleration engine is fully integrated with Pytorch framework. PyTorch models and software written with PyTorch API can run as-is, without any modifications.
Ampere Optimized ONNX Runtime
Ampere’s inference acceleration engine is fully integrated with ONNX Runtime framework. ONNX models and software written with ONNX Runtime API can run as-is, without any modifications.
Ampere Computing AI Docker Images
Ampere® Processors, with high performance Ampere Optimized Frameworks in Docker images, offer the best-in-class Artificial Intelligence inference performance for standard frameworks including TensorFlow, PyTorch and ONNXRT and llama.cpp. Ampere optimized containers come fully integrated with their respective frameworks.
Ampere Computing OCI Images
Ampere instances on OCI are some of the most cost-effective instances available in the Cloud today. OCI Ampere A1 started this with extremely high-performance shapes that with the ability to use the OCI Flex Shapes feature to provision at the single core resolution making this infrastructure very efficient.
Ampere Computing Azure Images
Azure series VMs in B, D and E series instances are available and offer some of the best price-performance for workloads on the Azure cloud. Ampere publishes 3 optimized frameworks for ease of access on Azure cloud marketplace, tested and proven AI inference for any model compatible with the framework.