NVIDIA H100 (SXM5, 80GB)
A single datacenter-grade AI chip. Sold to be installed in a server, not a home PC.
GPU memory
80 GB
GPU memory is the "workspace" a graphics card uses to hold an AI model while it runs. If a model is bigger than the available memory, it simply will not fit — like trying to fit a large book on a shelf that's too small. More in the wiki →
Power draw
700 W
Watts measure how much electricity something draws while it's running. More powerful hardware needs more electricity and produces more heat that has to be cooled. More in the wiki →
Ideal cooling
Passive - server airflow
The module has no fans of its own; the server's wall of high-pressure fans pushes air through its heatsink. It cannot run outside a proper server chassis. Running flat out, it turns 700 W of electricity into heat that has to go somewhere. We stock matching cooling gear →
Price
$30,000
This is a current typical market price, not a fixed list price — real prices move with supply and demand. NVIDIA has no public list price. The most-repeated industry estimate (Raymond James analyst research, cited widely since 2023) is $25,000–$40,000 per GPU. We use the middle of that range.
What this means in everyday terms
Running one of these draws as much power as about 0.6 average homes running around the clock. A full day of running one uses about 0.19 electric-car batteries worth of energy. An average home draws roughly 1,200 watts around the clock. This compares a build's power draw to that baseline. More in the wiki →
NVIDIA H100 (SXM5, 80GB) is sold to run inside a server or rack — it isn't something you plug into a home PC.
Where these numbers come from
Official NVIDIA spec page (memory, power). Price per The Decoder / Tom's Hardware reporting on Raymond James analyst estimates.