近日，芯片巨头英伟达（NVIDIA）在其官网上线了两款针对中国市场的消费级芯片产品，分别为英伟达RTX 5880 Ada工作站显卡和4090D显卡。相较于旗舰级产品RTX 6000，这两款显卡的单精度浮点性能降低了24%左右，整体性能减少了近1/4。
据了解，RTX 5880和4090D显卡在AI模型训练和推理方面的表现也有一定限制。行业人士透露，这两款显卡仅适用于70亿-650亿参数的模型训练，无法应对万亿级参数模型。而在推理方面，4090D显卡则能够稳定运行。
英伟达此举旨在应对美国禁令，维持在中国市场的竞争力。尽管性能有所下降，但两款显卡在消费级和专业领域应用中仍具有一定的市场前景。
Title: NVIDIA releases “China-specific” chips with nearly 25% performance reduction
Keywords: NVIDIA, China-specific chips, performance reduction
News content:
Recently, chip giant NVIDIA has launched two consumer-grade chip products specifically aimed at the Chinese market on its official website, namely the NVIDIA RTX 5880 Ada workstation graphics card and the 4090D graphics card. Compared to the flagship product RTX 6000, the performance of these two graphics cards in single-precision floating-point operations has decreased by about 24%, and their overall performance has reduced by nearly 1/4.
It is understood that RTX 5880 and 4090D graphics cards have certain limitations in terms of AI model training and inference. Industry insiders reveal that these two graphics cards are only suitable for training models with 700 million to 650 million parameters, and cannot handle trillion-level parameter models. In terms of inference, the 4090D graphics card can run stably.
NVIDIA’s move is aimed at maintaining its competitiveness in the Chinese market amidst US sanctions. Although the performance has been reduced, the two graphics cards still have market prospects in the consumer and professional fields.
【来源】https://www.tmtpost.com/6868819.html
Views: 1