Exxact Announces GPU-Accelerated Deep Learning Inference Platforms Now Shipping Worldwide

Exxact Corporation, a leading provider of high performance computing solutions (HPC) for GPU-accelerated deep learning research, announced that its NVIDIA T4-powered Deep Learning Inference Servers are now shipping worldwide

Exxact Deep Learning Inference Servers bring revolutionary multi-precision inference performance to efficiently accelerate the diverse applications of modern AI. Powered by NVIDIA T4 GPUs, Exxact Deep Learning Inference Servers are optimized for use in image and video search, video analytics, object classification and detection, and a host of other scenarios.

Read More: Decentralized AI Alliance Reaches 50 Members Worldwide

The NVIDIA T4 is the most versatile GPU to date bringing dramatic performance and efficiency gains to both deep learning training and inference. Built on NVIDIA’s Turing architecture, the T4 also includes Tensor Cores for acceleration of deep learning inference workflows.

“With the new NVIDIA T4, Exxact is focused on providing our customers with access to the most efficient accelerator solutions for AI inference workloads,” said Jason Chen, Vice President of Exxact Corporation. “As a leading provider of HPC and Deep Learning systems, Exxact continues to enhance the TensorEX server portfolio to help our customers drive every phase of the artificial intelligence life cycle and ultimately achieve their goals.”

Read More: Part 1: Can AI ML and IoT Solve the Job Crisis for Rural India?

Pre-installed with NVIDIA TensorRT™, Exxact Deep Learning Inference Servers maximizes data center throughput and GPU utilization, integrates with Kubernetes and Docker, and supports all popular AI models and frameworks such as TensorFlow, PyTorch, MXNet, Chainer, and Caffe2.

Exxact will offer various Deep Learning Inference Servers including:

• Entry-Level TensorEX 2U Server – Featuring 4x NVIDIA T4, 2x Intel Xeon Scalable (Silver), and 256GB memory
• Mid-Range TensorEX 2U Server – Featuring 8x NVIDIA T4, 2x Intel Xeon Scalable (Gold), and 512GB memory
• High-End TensorEX 4U Server – Featuring 20x NVIDIA T4, 2x Intel Xeon Scalable (Gold), and 512GB memory

All Exxact Deep Learning Inference and Training systems are fully turnkey — built to perform right out of the box and backed by a standard 3-year w******* and support service.

Read More: WISeKey announces at the Blockchain Summit in Bucharest the establishment of a WISeCoin Hub and Blockchain Center of Excellence in Romania