Huawei has decided to use GPGPU technology in its new generation artificial intelligence chips. With this architecture, which allows graphics processors to be used not only in image processing but also in operations requiring high computational power, the company aims to offer more effective solutions to the computational needs in the field of artificial intelligence.

Huawei will compete with Nvidia in the field of artificial intelligence

The GPGPU approach creates a serious performance advantage, especially in intensive data operations such as training and inference of large language models. The company’s move in this area has a strategic meaning within the current global balances.

Huawei, which is under the technology embargo imposed by the USA, continues to expand its area of ​​independence with alternative developments. It is stated that the company’s new chips are at a level that can compete with NVIDIA’s accelerators such as A100 and H100, which stand out in the field of artificial intelligence.

Among the applications where these chips can work with high efficiency, large language models, image recognition systems and deep learning algorithms stand out. Huawei is working with China’s largest semiconductor manufacturer SMIC in the production process of new generation artificial intelligence chips.

We can say that SMIC’s role in production is important, especially in terms of reducing external dependency. This cooperation sets an example for the development of domestic production capacity. At the same time, it reveals the point that China has reached in the process of establishing a more independent and competitive structure in the field of technology.

So what do you think about this issue? You can share your views with us in the comments section below.