uxlfoundation/oneDNN
oneAPI Deep Neural Network Library (oneDNN)
This library provides core computational building blocks for deep learning models, enabling applications like PyTorch or TensorFlow to run much faster. It takes common deep learning operations (like convolutions or matrix multiplications) and provides highly optimized versions. The primary users are developers of deep learning frameworks and applications who want to improve performance on various CPU and GPU hardware.
3,962 stars. Actively maintained with 242 commits in the last 30 days.
Use this if you are developing a deep learning framework or application and need to optimize its core computations for various Intel, AMD, or Arm processors and GPUs.
Not ideal if you are an end-user running deep learning models; instead, you should use a framework like PyTorch or TensorFlow that already incorporates this library.
Stars
3,962
Forks
1,110
Language
C++
License
Apache-2.0
Category
Last pushed
Mar 13, 2026
Commits (30d)
242
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/uxlfoundation/oneDNN"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
apache/tvm
Open Machine Learning Compiler Framework
Tencent/ncnn
ncnn is a high-performance neural network inference framework optimized for the mobile platform
OpenMined/TenSEAL
A library for doing homomorphic encryption operations on tensors
iree-org/iree-turbine
IREE's PyTorch Frontend, based on Torch Dynamo.
ARM-software/armnn
Arm NN ML Software.