GPU Cloud Server

Product Specifications

2026-02-24 06:43:19

GPU Computing Accelerated

GPU computing accelerated GPU cloud servers use the GPU pass-through technology and are mainly suitable for AI deep learning & training, inference, scientific computing, video transcoding, and graphics processing.

Commercially available: PI7,PI2

Features of computing accelerated GPU cloud server

Specification Name

Graphics Card Specification

Quantity of Graphics Cards

Single-GPU Performance

Disk Type

PI7

Nvidia Tesla A10

1, 2, 4

31.2 TFLOPS single-precision floating point computing
62.5 TFLOPS TF32AI acceleration

Common I/O
High I/O
Generic SSD
Ultra-high I/O

PI2

Nvidia Tesla T4

1, 2, 4

8.1 TFLOPS single-precision floating point computing
130 TOPS INT8 computing
260 TOPS INT4 computing

Common I/O
High I/O
Generic SSD
Ultra-high I/O

GPU image acceleration basic type

The GPU cloud server of GPU image acceleration basic type is based on the NVIDIA GRID virtual GPU technology. A public image is integrated with the GRID drive and contains the license for NVIDIA GRID vWS to effectively reduce the cost of use in small-scale demands. This kind of cloud server is suitable for image rendering and small-scale AI inference scenarios.

Commercially available: G7,

Features of GPU cloud server of image acceleration basic type

Specification Name

Graphics Card Specification

Quantity of Graphics Cards

Single-GPU Performance

Disk Type

G7

Nvidia Tesla A10

1/4, 1/2, 1

31.2 TFLOPS single-precision floating point computing
62.5 TFLOPS TF32AI acceleration

Common I/O
High I/O
Generic SSD
Ultra-high I/O

 


c4qj3FnYhwtD