For years, China’s domestic chip developers have viewed Nvidia with admiration, as the US semiconductor giant has established undisputed dominance in the industry through its hardware capabilities and proprietary CUDA toolkit, which engineers use to develop applications on the firm’s graphics processing units (GPUs).
Following the launch of DeepSeek’s V3 and R1 models, domestic chip developers are now rallying to break away from Nvidia’s orbit.
The implications of DeepSeek’s AI model breakthrough for Nvidia and other US tech titans have been heatedly talked about in the tech industry over the past weeks.
Precisely how many chips were used by the Hangzhou-based company to train its V3 and R1 models remains a matter of debate. There is speculation, however, that chips from Huawei Technologies played a role in DeepSeek’s successful development of powerful, low-cost AI models.
Source link
