Source: wikibot/half-precision-floating-point-format
= Half-precision floating-point format
{wiki=Half-precision_floating-point_format}
Half-precision floating-point format, often referred to as "half" or "binary16," is a computer number format that occupies 16 bits (2 bytes) in memory and is typically used for representing floating-point numbers with lower precision and range than single-precision (float) or double-precision (double) formats.