Installation

  1. Install from Pypi

  2. Install from Source

    2.1. Prerequisites

    2.2. Install Intel Extension for Transformers

  3. System Requirements

    3.1. Validated Hardware Environment

    3.2. Validated Software Environment

Install from Pypi

Binary builds for python 3.8, 3.9 and 3.10 are available in Pypi

Note: Recommend install protobuf <= 3.20.0 if use onnxruntime <= 1.11

# install stable basic version from pypi
pip install intel-extension-for-transformers

# install nightly version
pip install -i https://test.pypi.org/simple/ intel-extension-for-transformers
# or install nightly version with only backend
pip install -i https://test.pypi.org/simple/ intel-extension-for-transformers-backend

# install stable basic version from from conda
conda install -c intel intel_extension_for_transformers

Install from Source

Prerequisites

The following prerequisites and requirements must be satisfied for a successful installation:

  • Python version: 3.8 or 3.9 or 3.10

  • GCC (on Linux) or Visual Studio (on Windows)

Install Intel Extension for Transformers

git clone https://github.com/intel/intel-extension-for-transformers.git itrex
cd itrex
pip install -r requirements.txt
# Install intel_extension_for_transformers
pip install -v .

System Requirements

Validated Hardware Environment

Intel® Extension for Transformers supports systems based on Intel 64 architecture or compatible processors that are specifically optimized for the following CPUs:

  • Intel Xeon Scalable processor (formerly Cascade Lake, Icelake)

  • Future Intel Xeon Scalable processor (code name Sapphire Rapids)

Validated Software Environment

  • OS version: CentOS 8.4, Ubuntu 20.04

  • Python version: 3.8, 3.9, 3.10

Framework Intel TensorFlow PyTorch IPEX
Version 2.12.0
2.11.0
2.0.0+cpu
1.13.1+cpu
1.12.0+cpu
2.0.0
1.13.0

  • OS version: Windows 10

  • Python version: 3.8, 3.9, 3.10

Framework Intel TensorFlow PyTorch
Version 2.9.1
2.0.0+cpu