Skip to content

Glossary: Serving LLMs at Scale

FP8

An 8-bit floating-point number format. It is the default quantization choice on modern NVIDIA GPUs and nearly free in quality; INT8 is the 8-bit integer alternative for older GPUs, and NVFP4 is a 4-bit format for Blackwell.

Where the book teaches it

Related terms

QuantizationW4A16 / W8A8BF16