I've been working on a project where I'm trying to collect and model the behavior of every floating-point format I can get my hands on.

The AVX10.2 ISA (Intel/AMD) appears to include conversion instructions for the quarter-precision types E5M2 and E4M3 described in the “OCP 8-bit Floating Point Specification”.

Also, one might be interested in takum arithmetic <https://github.com/takum-arithmetic/libtakum> , which also had a paper presented at ARITH 2025 <https://www.arith2025.org/program.html> .