AI workloads
Precision (FP8, BF16, FP16)
The number format used for calculations. Lower precision such as FP8 uses less memory and runs faster, at some cost in numerical accuracy.
Why it matters
Precision changes memory needs and speed, so comparing GPUs requires knowing which format a claimed performance figure uses.
Want a deeper answer?
Ask Kova to explain Precision (FP8, BF16, FP16) in the context of your workload. Meet Kova →