NNCF conversion and optimum-cli tool
#1
by
Echo9Zulu
- opened
Hello! I am working on learning more about the openvino conversion process and see these models referenced in the src. Can you explain a bit about how this works and maybe provide some documentation?
I am trying to diagnose some performance issues with openvino on CPU only; llama.cpp is eating our lunch, though in this this issue I tested against full precision torch: https://github.com/huggingface/optimum-intel/issues/1275