Partner Introduction

Zhongshan T-Chip Technology Official Website

   Zhongshan T-Chip Technology (hereinafter referred to as" T-Chip "or" the company ") was established in 2005. It is a national high-tech enterprise, Guangdong Internet+ enterprise and Zhongshan engineering technology center. It is an industry-leading service provider for intelligent hardware design, manufacturing and AI application solutions.

   T-chip has focused on the development of intelligent hardware for more than 10 years and has conducted in-depth cooperation with frontline AI chip manufacturers such as Rockchip, Huawei Hisilicon, BITMAIN, etc. to meet the development needs of AI development. It has focused on the AI edge computing field, and launched AI edge computing modules, AI edge computing boxes, edge intelligent servers and AIOT gateways, providing all-round AI edge computing products and solutions for the urban brain, intelligent security, medical care, city, transportation, construction site and fire protection.

Product Introduction

AI 1U Cluster Server

   Cluster Server R1A AI cluster server is equipped with 8 Core-1684JD4 computing modules + 1 Core-3588SJD4 main module, with an aggregate computing power of 140.8Tops and a total memory of 104GB. It supports mainstream deep learning frameworks such as cafe / tensorflow / mxnet / pytorch / PaddlePaddle. It integrates 4 Gigabit Ethernet, expands 4G LTE, and has HDMI, USB2.0 and other interfaces. It supports Ubuntu, Debian, Linux and other operating systems; The standard 1U rack server can be applied to the fields of urban brain, intelligent security, medical treatment, city, transportation, construction site and intelligent fire protection.

Specifications

Product Model
Cluster Server R1
Application Scenarios

Urban brain, intelligent security, medical treatment, city,  transportation,  construction site and  fire control

AI Chip
SOPHON BM1684
AI Performance
140.8T@INT8,with 8 computing modules(8 * Core-1684-JD4 computing module)
CPU
1 * Core-3588S-JD4 main module
Memory
96GB LPDDR4X
Video decoding capability
Support 256 H.265/H.264 1080p video decoding
Video encoding capability
1080p@400fps
Picture Codec Capability
MJPEG image encoding and decoding up to 1080P@3840fps
Storage
256GB eMMC
Network Interface
4 * integrated Ethernet
USB Interface
2 * USB2.0
Display Interface
HDMI
Wireless Network
Optional 4G LTE/5G expansion
Size
1U standard
Other Parameters

1 * HDMI , 1 * TF, SATA, 4G LTE, 5G, standard 1U rack mount, computing core board optional for configuration.

Operating System
Support Android, Ubuntu, Linux OS
Deep Learning Framework
Support caffe/tensorflow/mxnet/pytorch/PaddlePaddle and other deep learning frameworks
Intelligent Algorithm

Support face detection, tracking, recognition, pose recognition, helmet detection, etc