- Nvidia’s H100 graphics card sells for over $40,000 on eBay.
- High-end chips remain essential for training and deploying AI software.
- The price was pointed out by 3D gaming pioneer and former Meta Consulting Technology Chief John Carmack on Twitter.
Jen-Hsun Huang, President and CEO of Nvidia Corp., unveiled the EGX edge supercomputing platform at the company’s event at Mobile World Congress Americas in Los Angeles, California on October 21, 2019. Announced.
Patrick T. Fallon | Bloomberg | Bloomberg | Getty Images
With soaring demand for the chips needed to train and deploy artificial intelligence software, Nvidia’s cutting-edge graphics cards sell for over $40,000 on eBay.
The price of Nvidia’s H100 processor was pointed out by 3D gaming pioneer and former Meta Consulting tech chief John Carmack. on TwitterOn Friday, at least eight H100s were listed on eBay for prices ranging from $39,995 to just under $46,000. Some retailers have offered it for around $36,000 in the past.
The H100, announced last year, is Nvidia’s latest flagship AI chip and the successor to the A100, a roughly $10,000 chip that has been dubbed a “workhorse” for AI applications.
Developers use H100 to build so-called Large Language Models (LLMs), which are central to AI applications such as OpenAI’s ChatGPT. These systems are expensive to run and require powerful computers to process large amounts of terabytes of data for days or weeks at a time. It also relies on massive computational power to enable AI models to generate text, images, or predictions.
Training AI models, especially large models like GPT, requires hundreds of high-end Nvidia GPUs working together.
Microsoft spent hundreds of millions of dollars on tens of thousands of Nvidia A100 chips to help build ChatGPT. Nvidia dominates the majority of the AI chip market.
Nvidia also offers a supercomputer with eight GPUs working together called DGX. Earlier this year, the company announced a new service that allows companies to rent access to his DGX computer for $37,000 a month. For that price, the system uses his Nvidia’s old A100 processor.
According to Nvidia, the H100 is the first chip optimized for the specific AI architecture behind many of the recent AI advancements called Transformers. Industry experts say more powerful chips will be needed to build models that are even larger and require more data than those currently available.
clock: Google chip manufacturing takes aim at Nvidia’s generative AI
