Scenarios

Release Notes

Discontinuing the vGPU-based GN7 GPU Computing Instance Specifications

Upgrading GRID Driver

Decommissioning Some Tesla Driver Versions from the Automatic Driver Installation Feature

Announcements

Release Notes and Announcements

Overview

Strengths

Notes

Product Introduction

Computing Instance

Rendering Instance

Instance Types

Billing Overview

Renewal

Billing

Getting Started

Logging In to Instances

Restarting Instances

NVIDIA Driver Installation Guide

Installing Tesla Driver

Installing GRID Driver

Installing CUDA Driver

Installing NVIDIA Driver

Uninstalling Tesla Driver

Uninstalling GRID Driver

Uninstalling NVIDIA Driver

Upgrading Tesla Driver

Upgrading GRID 16 Driver

Upgrading NVIDIA Driver

Using GPU Monitoring and Alarm

User Guide

Installing NVIDIA Container Toolkit on a Linux Cloud GPU Service

Using Windows Cloud GPU Service to build a Deep Learning Environment

Implementing Image Quality Enhancement with GN7vi Instances

Using Docker to Install TensorFlow and Set GPU/CPU Support

Using GPU Instance to Train ViT Model

Use Cases

GPU Instance Troubleshooting Guide

Troubleshooting Common Xid Errors

Collecting Log for GPU Instances

GPU Usage Shows 100%

VNC Login Failures

Troubleshooting

FAQs

Special Terms for Committed Sales Model

Related Agreement

Contact Us

Buy Now

https://buy.tencentcloud.com/cvm

https://www.tencentcloud.com/document/product/560/75945

Product Pricing

https://intl.cloud.tencent.com/pricing/cvm/overview

Introduces the advantages of Cloud GPU Service compared to self-built GPU servers, helping you quickly understand Cloud GPU Service.

https://www.tencentcloud.com/document/product/560/8015

Cloud GPU Service deliver exceptional parallel computing capabilities, making it ideal for scenarios such as AI training, AI inference, scientific computing, cloud gaming, cloud rendering, and video encoding/decoding.

https://www.tencentcloud.com/document/product/560/76772

Cloud GPU Service offers multiple billing methods, allowing you to flexibly choose based on your business scenarios to optimize your costs.

Billing Modes

https://www.tencentcloud.com/document/product/560/75944

Introduces the related concepts, use cases, billing methods, and other information of Tencent Cloud's Cloud GPU Service to help you better understand the product.

Learning About Cloud GPU Service

Cloud GPU Service provides diverse instance types to meet the needs of various business use cases. You can select the appropriate instance type based on your actual needs.

Instance Specifications

https://www.tencentcloud.com/document/product/560/11625

Tencent Cloud provides multiple driver installation methods, allowing you to select the appropriate driver type and installation approach based on your business scenarios.

Driver Installation

https://www.tencentcloud.com/document/product/560/75947

Through the monitoring and alarm feature, you can proactively identify potential failures or security risks that may arise during the operation of Cloud GPU Service.

Monitoring and Alarms

https://www.tencentcloud.com/document/product/560/75956

Introduces how to configure and use Cloud GPU Service instances, GPU drivers, and other core capabilities, and covers daily Ops management operations to help you get started.

Using Windows Cloud GPU Service to set up a development environment suitable for deep learning.

Deep Learning Environment Setup

https://www.tencentcloud.com/document/product/560/75962

Use the video encoding/decoding feature and AI quality enhancement feature of the video-enhanced instance GN7vi to complete video quality processing.

Video Quality Enhancement

https://www.tencentcloud.com/document/product/560/50604

Using Cloud GPU Service for offline training of ViT models to accomplish simple image classification tasks.

Training the ViT Model

https://www.tencentcloud.com/document/product/560/50233

Learn about the best practices for Cloud GPU Service to help you use Cloud GPU Service more efficiently.

Cloud GPU Service is an elastic computing service that provides GPU compute capability and features exceptional parallel computing power. As a key enabler at the IaaS layer, it serves scenarios such as generative AI, autonomous driving, deep learning training, scientific computing, graphics and image processing, and video encoding/decoding. Tencent Cloud offers readily available compute capability at any time, effectively alleviating your computational pressure and enhancing business efficiency and competitiveness.

Cloud GPU Service

The GN10Xp and GT4 GPU computing instances are equipped with V100 and A100 GPUs, respectively. They feature powerful double-precision floating-point computing capabilities and 300–600 GB/s NVLink bandwidth, making them ideal for large-scale AI training acceleration.

The PNV4 and GN7 GPU computing instances are equipped with A10 and T4 GPUs, respectively. They provide high performance-per-watt single-precision floating-point computing capabilities, making them widely used in online and offline inference scenarios.

The GN10Xp and GT4 GPU computing instances are equipped with V100 and A100 GPUs, respectively. They feature large video memory, high computing power, and low-latency interconnects between cards, providing large-scale parallel computing capabilities for scientific computing scenarios such as computational fluid dynamics (CFD) , seismic analysis, and genomics.

The GNV4v and GN7vw GPU rendering instances are equipped with NVIDIA A10 and T4 GPUs, respectively, and feature GRID drivers optimized for cloud gaming. They support 1/2 and 1/4 GPU slicing, aimed at optimizing the graphics cost for single-stream games while ensuring high image quality and low latency.

The GA3 instance features the Tencent Cloud Star Lake GA01 and offers fine-grained 1/6 slicing. Specifically designed for cloud gaming, it provides a highly cost-effective solution for the industry.

The GNV4 and GN7vw GPU rendering instances are equipped with A10 and T4 GPUs, respectively. Featuring GRID drivers optimized for rendering scenarios, they enable rapid image and video rendering, improving online and offline rendering efficiency for industries such as film, animation, and industrial design.

The video-enhanced GN7vi instances, powered by T4 GPUs, are equipped with Tencent’s self-developed Media Video Fusion AI technology, featuring a TSC codec engine along with a quality enhancement toolkit. These instances are ideal for video-on-demand (VOD) and live streaming scenarios.

tencent cloud

AI Training

AI Inference

Scientific Computing

Cloud Gaming (GS)

Cloud Rendering

Video Encoding and Decoding