Nvidia will launch a new artificial intelligence chipset for China at a significantly lower price than its recently restricted H20 model. It plans to start mass production as early as June. The GPU or graphics processing unit will be part of Nvidia’s latest generation Blackwell-architecture AI processors and is expected to be priced between $6,500 and $8,000.
Interestingly, the chipmaker is expected to begin mass production of these cheaper chips in June. The proposed chip will have weaker specifications compared to the current H20 models and will be produced using simple manufacturing techniques. Nvidia has said that the new chip restrictions will result in a $5.5 billion inventory write-off and the loss of nearly $15 billion in potential sales. Moreover, the chip will use traditional GDDR7 memory to circumvent restrictions on exporting chips with high bandwidth memory (HBM), which is used in more advanced models.
The slightly toned down chip will be based on Nvidia’s older RTX Pro 6000D processor, a server-class graphics processor. Finally, Nvidia will not use Taiwan Semiconductor Manufacturing’s (TSM) CoWoS (Chip-on-Wafer-on-Substrate) packaging technology for these chips.
Nvidia is exploring different strategies to continue exporting chips to China, which remains its second-largest market. This is the third time the chip giant is changing chip specifications to comply with the U.S.’ chip export restrictions.
See What’s Next in Tech With the Fast Forward Newsletter
Tweets From @varindiamag
Nothing to see here - yet
When they Tweet, their Tweets will show up here.
