Nvidia plans to launch the B100AI accelerator with a power consumption of up to 1000W, which represents a 42% increase compared to previous products. A Dell executive stated that the power consumption of GPUs will continue to rise, but direct liquid cooling may no longer be an essential solution. Nvidia's next-generation GPUs may include the B200 Superchip, integrated with the Grace CPU, with power consumption potentially reaching around 1300 watts.

["B100's speed could exceed H100's by more than 4 times, featuring a chip combination design", "GPT-5 may achieve general artificial intelligence, requiring substantial computational resources", "Microsoft has invested over $10 billion into OpenAI", "NVIDIA's B100 and OpenAI's GPT-5 are anticipated to launch in 2024", "Industry experts predict both will lead to significant improvements in computing speed and intelligence"]

NVIDIA plans to advance the release date of the Blackwell B100 GPU to Q2 2024. NVIDIA has selected SK Hynix's HBM3e memory for the Blackwell B100 GPU. The early launch of the Blackwell B100 is aimed at meeting the rapidly growing demand for AI. NVIDIA received samples of SK Hynix's HBM3e memory in August of this year. The Blackwell B100 will help NVIDIA maintain its leading position in the AI market.