The IEEE 754 standard defines the representation and the properties of the floating-point numbers used as surrogates for reals in computer programs. Most programming languages only support the 32-bit (Float32) and 64-bit (Float64) formats implemented in hardware. Machine learning, computer graphics, and numerical algorithms analysis all have a need for smaller formats, which are often neither supported in hardware, nor are they available as established types in programming languages. The MicroFloatingPoints.jl Julia library offers a parametric type that can be instantiated to compute with IEEE 754-compliant floating-point numbers with varying ranges and precisions (up to and including Float32). It also provides the programmer with various means to visualize what is computed.