DeepSeek reportedly developing its own AI inference chip, aiming to reduce reliance on Nvidia and Huawei

CoinFeed July 7 news, according to Reuters, Chinese startup DeepSeek is developing its own chip for the AI inference stage to reduce reliance on Nvidia and Huawei chips. Sources say the project started about a year ago, and DeepSeek has recently quietly increased recruitment of chip design engineers and is in talks with multiple chip design, foundry, and storage companies. If progress goes smoothly, DeepSeek will shift from focusing on model R&D to semiconductor layout, building its own inference chips to meet growing computing power demand, while seeking alternatives to Nvidia's high-end products under U.S. export controls, and competing with Chinese tech companies like Huawei, Alibaba, and Baidu that are already developing their own AI chips.