Home >Technology peripherals >AI >RX 7900 XT is now compatible with ROCm 5.7 and supports PyTorch ML and AI acceleration

RX 7900 XT is now compatible with ROCm 5.7 and supports PyTorch ML and AI acceleration

PHPz
PHPzforward
2023-11-22 20:43:23868browse

AMD released the ROCm 5.7.1 driver update for Ubuntu Linux last month, adding support for ROCm and PyTorch to Radeon Pro W7900 and Radeon RX 7900 XTX graphics cards. Ahead of the upcoming AMD Advancing AI event, AMD announced that it will provide ROCm and PyTorch support for the RX 7900 XT graphics card, which also uses the RDNA 3 architecture. This will cover all top products using the RDNA 3 architecture

RX 7900 XT现已兼容ROCm 5.7,支持PyTorch ML和AI加速

Erik Hultgren, AMD software product manager, said: “We are very excited about the latest products in our portfolio. These high-end GPUs combined with ROCm make the use of artificial intelligence easier from both a software and hardware perspective. Therefore, developers can choose the most suitable solution based on their needs. RX 7900 XT has 20GB of large video memory and a total of 168 AI acceleration units on the Navi 31 chip, which can provide fast machine learning training and inference performance. In addition, this GPU supports machine learning models and algorithms using PyTorch. These models and algorithms have been transcoded to be optimized for AMD and are compatible with computers running Ubuntu Linux system version 22.04.3

RX 7900 XT现已兼容ROCm 5.7,支持PyTorch ML和AI加速

However, currently AMD still has many RDNA 3 architecture GPUs that cannot support related applications, so many consumers hope to get ROCm support. In contrast, NVIDIA has always provided better general computing support on consumer-grade graphics cards, and CUDA has also become a consideration for many consumers when purchasing graphics cards. Compared with this, AMD's ROCm support is still quite limited and is being continuously optimized

The above is the detailed content of RX 7900 XT is now compatible with ROCm 5.7 and supports PyTorch ML and AI acceleration. For more information, please follow other related articles on the PHP Chinese website!

Statement:
This article is reproduced at:sohu.com. If there is any infringement, please contact admin@php.cn delete