How Huawei’s New AI Chip Aims to Rival Nvidia’s GPUs
Huawei is developing a new AI chip that functions more like a general‑purpose GPU, aiming to match Nvidia and AMD architectures and simplify the transition for Chinese AI developers, while still facing challenges such as adapting CUDA‑based software and overcoming export restrictions.
According to a report in The Information, Huawei Technologies is developing a new artificial‑intelligence chip design that will boost processor performance and handle a broader range of AI development tasks.
The design is intended to be closer to the architectures used by Nvidia and AMD, essentially a general‑purpose GPU rather than a specialized AI chip, which should make it easier for Chinese AI developers to transition from Nvidia hardware.
Huawei aims to make its chip more attractive to Chinese technology companies despite increasing difficulty in obtaining Nvidia processors due to U.S. export controls, although Nvidia remains the preferred choice.
Chinese tech firms have tested Huawei’s chips but few have fully adopted them, largely because developers have relied on Nvidia’s CUDA software ecosystem for years, creating compatibility challenges.
The planned new chip will allow AI developers using Huawei processors to run code written for Nvidia’s software platform, but Huawei still needs to develop software that translates Nvidia instructions into a form its own chips can understand.
Source: EETOP
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Architects' Tech Alliance
Sharing project experiences, insights into cutting-edge architectures, focusing on cloud computing, microservices, big data, hyper-convergence, storage, data protection, artificial intelligence, industry practices and solutions.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.