Nvidia H100 GPU prices surge 30% as AI compute demand holds firm
Key Summary
Nvidia's H100 GPU has seen a 30% increase in value over the past year, driven by steady demand for AI compute, according to a report. The H100, launched in 2022, continues to hold its value despite the introduction of newer chips, with prices ranging from $25,000 to $40,000 on the secondary market.
Precedent & Market Context
The growth rate of training compute for modern AI models has been approximately 5x per year since 2020.
A recent report by Epoch AI estimated that the global AI chip count is projected to double every nine months, potentially reaching 200 million by the end of 2028.
As AI compute matures into a tradable asset class, market expectations for rental rates are shifting.
CME Group has introduced non-tradable forward curve pricing data for GPU rental indices, covering H100 and B200 chips.
The data reflects market expectations for rental rates to soften as more capacity comes online.
Market Analysis
The median H100 rental rates sit between $3.3 and $3.6 per GPU-hour as of mid-October 2026.
The median on-demand rate stood at $3.49 per GPU-hour as of October 16, 2026.
Advertisement
The H100 is built on Nvidia's Hopper architecture and launched in Q4 2022 as the company's flagship data center product.
The chip carries 80GB of HBM3 memory, delivers up to 989 TFLOPS of FP16 tensor performance, and draws a 700W TDP.
Recent price hikes from service providers, combined with persistent interest in AI applications, point to pressure on compute costs for end users.
Organizations investing in AI infrastructure may face higher operating costs in the near term.
The variation across indices matters. A 13% gain and a 79% gain describe very different markets.
Anyone making decisions on H100 pricing should look at which benchmark they are using and what period it compares against.
The gap between hyperscaler and specialist pricing is another variable.
A 1.7x to 2.7x premium gives cost-conscious buyers a reason to shop around.