ZenDNN (Zen Deep Neural Network) Library accelerates deep learning inference applications on AMD CPUs. This library, which includes APIs for basic neural network building blocks optimized for AMD CPUs, targets deep learning application and framework developers with the goal of improving inference performance on AMD CPUs.

ZenDNN v3.2 Highlights                      Downloads   Documentation

  • Enabled, tuned, and optimized for AMD 2nd and 3rd Gen EPYC processors 
  • Integrated with TensorFlow v2.7, PyTorch v1.9.0, and ONNXRT v1.8.0
  • Features highly optimized primitives for AMD CPUs, targeting a variety of workloads, including computer vision, natural language processing, and recommender systems
  • Tested with a variety of neural network models across three major workload types:
    • Computer Vision: AlexNet, InceptionV3, InceptionV4, GoogLeNet, ResNet50, ResNet152, VGG16, and VGG19
    • Natural Language Processing: BERT
    • Recommender Systems: DLRM, Wide & Deep
  • Supported on Ubuntu 18.04, Ubuntu 20.04, RHEL 8.3, and CentOS 8.3
ZenDNN will be available shortly on GitHub (https://github.com/amd/ZenDNN). ZenDNN library is intended to be used in conjunction with the frameworks mentioned above and cannot be used independently.

Previous Release

ZenDNN v3.1 Highlights

  • Enabled, tuned, and optimized for AMD 2nd and 3rd Gen EPYC processors 
  • Integrated with TensorFlow v2.5 
  • Features highly optimized primitives for AMD CPUs, including: Convolution, MatMul, Elementwise (Binary), BatchNorm, and ReLU 
  • Tested with a variety of neural network models across three major workload types:
    • Computer Vision: AlexNet, InceptionV3, InceptionV4, GoogLeNet, ResNet50, ResNet152, VGG16, and VGG19
    • Natural Language Processing: BERT
    • Recommender Systems: DLRM, Wide & Deep
  • Supported on Ubuntu 18.04, Ubuntu 19.10, Ubuntu 20.04, RHEL 8.3, and CentOS 8.3
  • INT8 models supported and tested with RESNET50 variants

Resources and Technical Support

Resources

Documentation

ZenDNN User Guide

TensorFlow + ZenDNN User Guide

PyTorch + ZenDNN User Guide

ONNXRT + ZenDNN User Guide

ZenDNN support matrix lists the details of ZenDNN releases and other relevant artifacts.

Technical Support

Refer Technical Support for support options.

Downloads

TensorFlow and PyTorch are available with different Python versions. ONNXRT is available with Python v3.8 and docker.

For previous releases of ZenDNN, please refer here.

File Name Version Size Launch Date OS Bitness Description
TensorFlow

File Name

Version

3.2

Size

206 MB

Launch Date

12/21/2021

OS

Ubuntu/ RHEL

Bitness

64-bit

Description

TF v2.7 + ZenDNN binary release package with python v3.6. MD5 Checksum: ff5fc72098e4703cd6c2cfa705025bb9

File Name

Version

3.2

Size

206 MB

Launch Date

12/21/2021

OS

Ubuntu/RHEL

Bitness

64-bit

Description

TF v2.7 + ZenDNN binary release package with python v3.7. MD5 Checksum: c009a285b574a0c25d9c21b25d663ffe

File Name

Version

3.2

Size

206 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

TF v2.7 + ZenDNN binary release package with python v3.8. MD5 Checksum: 6dc3ab78fffc9927baa77a60c3bf3be8

File Name

Version

3.2

Size

206 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

TF v2.7 + ZenDNN binary release package with python v3.9. MD5 Checksum: b1ec86e2eaa264b198a5699524d20492

File Name

Version

3.2

Size

123 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

TF v2.7 + ZenDNN binary release package for C++ API interface. MD5 Checksum: ff65a75cb513fdfe1323ac70edda25af
PyTorch

File Name

Version

3.2

Size

115 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

PT v1.9.0 + ZenDNN binary release package with python v3.6. MD5 Checksum: 44079e976374eb85a3db30de4d44e069

File Name

Version

3.2

Size

115 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

PT v1.9.0 + ZenDNN binary release package with python v3.7. MD5 Checksum: 3000feb8b73671d44a30b51684ef4cb4

File Name

Version

3.2

Size

115 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

PT v1.9.0 + ZenDNN binary release package with python v3.8. MD5 Checksum: fcef8e8bbf459786bab820e398d94f94

File Name

Version

3.2

Size

115 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

PT v1.9.0 + ZenDNN binary release package with python v3.9. MD5 Checksum: 5b318e8a6b3f6fc2dd96814fe1975f9b
ONNXRT

File Name

Version

3.2

Size

109 MB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

ONNXRT v1.8.0 + ZenDNN binary release. MD5 Checksum: afb9ca63bf416d37875055b670f79953

File Name

Version

3.2

Size

8.83 GB

Launch Date

12/21/2021

OS

Ubuntu and RHEL

Bitness

64-bit

Description

ONNXRT v1.8.0 + ZenDNN docker image release package. MD5 Checksum: 6db047b9cf005b69a8b2bfa56713156b