Skip to content

Latest commit

 

History

History
7 lines (5 loc) · 486 Bytes

File metadata and controls

7 lines (5 loc) · 486 Bytes

Running PyTorch model using IPEX-LLM on Intel CPU

This folder contains examples of running any PyTorch model on IPEX-LLM (with "one-line code change"):

  • Model: examples of running PyTorch models (e.g., Openai Whisper, LLaMA2, ChatGLM2, Falcon, MPT, Baichuan2, etc.) using INT4 optimizations
  • More-Data-Types: examples of applying other low bit optimizations (NF4/INT5/INT8, etc.)
  • Save-Load: examples of saving and loading low-bit models