Glossary — plain words only
For smart people who never bought hardware. No jargon survives this page.
GPU memory (VRAM)
GPU memory is the workbench — the whole model must fit here or it doesn't run at all. Regular RAM cannot substitute: it sits on a link far too slow for the GPU, which would spend nearly all its time waiting instead of computing.
Memory bandwidth
How fast the GPU re-reads the model from its own memory. To generate every single word, the GPU re-reads the ENTIRE model — all of it, every word. So faster memory means faster answers: double the bandwidth and the model talks roughly twice as fast.
Watts (power draw)
Electricity burned every second the machine runs — this is a real monthly bill, not a spec-sheet footnote. We translate it below into homes and electric-car batteries so you can feel the size of it. An average home draws ~1,200 W around the clock; a typical electric-car battery holds ~90 kWh. That's why every product here shows "homes" and "EV batteries per day".
Parameters
Parameters are the size of the model's brain — more parameters usually means a smarter model, but every one of them has to live in GPU memory.
CPU
The CPU is the traffic director — it does almost none of the AI math. Don't overspend here; the GPU does the work.
Quantization
Quantization means storing the model's numbers less precisely so it fits in less memory — 8-bit halves it, 4-bit halves it again, with some quality loss.
Cluster
A cluster is several machines wired together so they can act like one bigger machine. The honest part: real datacenter clustering works because NVLink and InfiniBand let GPUs share memory at enormous speed. Eight desktop cards in a closet do NOT make a datacenter — the links between them are so slow that a big model split across them runs painfully slowly. Desktop cards are for models that fit on ONE card.
HBM vs GDDR
Two kinds of GPU memory. GDDR is the affordable kind on desktop cards; HBM is stacked right next to the chip and is several times faster — that's a big reason datacenter GPUs answer so much faster (remember: the GPU re-reads the whole model for every word).