BINARY_FLOAT is a 32-bit, single-precision, floating-point number data type. Each BINARY_FLOAT value requires 5 bytes, including a length byte. See the information in the note provided for the FLOAT data type for more details about floating-point numbers.