eBay is now offering high-end Nvidia graphics cards, but you can still expect to break the bank if you’re looking to get one. The successor to the AI-powered ChatGPT chip, known as the H100, can be seen listed on eBay for over $40,000. In some cases, the cost is slightly lower, but the highest markup seen is around $46,000.
This jaw-dropping price tag is significantly higher than the $10,000 cost set by the previous model, the A100. The driving factor behind this rise in cost is the increased demand for AI technology, which skyrocketed thanks to the breakout success of OpenAI’s ChatGPT. Building such models requires processing huge datasets, and this involves running hundreds of powerful GPUs simultaneously for days or even weeks on end.
Showing its commitment to AI development, Microsoft invested millions of dollars in equipping the Azure cloud computing infrastructure with A100s. In addition, Nvidia is offering access to its specialised DGX supercomputer for $37,000 a month, which is also powered by the A100s.
And now, the H100 has been designed to offer capabilities specifically tailored to AI applications. Nvidia calls this their crowning achievement and hopes that the new chip will usher in an all-new era of powerful AI systems, which seems more likely now that even Microsoft has started reaping the rewards of the increasingly pervasive AI technology.
Nvidia Corporation is a world-renowned American multinational technology company and market leader in artificial intelligence computing, graphics processing unit manufacturing and programmable shader modelling. Founded by visionary entrepreneurs Jensen Huang, Chris Malachowsky and Curtis Priem in 1993, Nvidia has played a pivotal role in the evolution of gaming and AI technology.
Jensen Huang is the chief executive officer, co-founder and president of Nvidia. He has held leadership positions at both Nvidia and LSI Logic, and is renowned for his work in pioneering the core technologies used in the latest artificial intelligence advancements. Huang is most well-known for his development of the CUDA architecture, which is an industry milestone for harnessing parallel computing through general-purpose graphics processing units.