fix(cpu): preserve raw fp16 bits in x86 quantize conversion - #685

Open

kkkzbh wants to merge 3 commits into

UbiquitousLearning:mainfrom

kkkzbh:codex/fix-x86-fp16-quantize

Commits on Jun 13, 2026

fix(cpu): preserve raw fp16 bits in x86 quantize
kkkzbh
committed
8647bda
docs(cpu): explain x86 fp16 bit helpers
kkkzbh
committed
7011227

Commits on Jun 26, 2026

Merge branch 'main' into codex/fix-x86-fp16-quantize
chenghuaWang
authored
bb82cb7