The Deep Diamond library provides fast tensor and neural network computations in Clojure using native CPU and GPU routines.
Open source
24 alternativesoneDNN
oneDNN provides a library of optimized deep neural network operations for oneAPI.
About oneDNN
oneDNN is an open-source, cross-platform library of deep learning operations, including matrix multiplication, convolution, attention and normalization. It serves as an acceleration layer for frameworks such as PyTorch, TensorFlow, OpenVINO and ONNX Runtime.
The library is optimized for Intel 64/AMD64 processors, Arm AArch64 processors and Intel Graphics, with a single API covering those devices. Support for NVIDIA GPU, AMD GPU, OpenPOWER PPC64, IBMz s390x and RISC-V is described as experimental.
Key features
- Matrix multiplication, convolution, attention and normalization operations
- Single API targeting x64, AArch64, RISC-V CPUs and Intel GPUs
- Runtime hardware detection with JIT-generated kernels for CPU features
- Low-precision data types including f32, bf16, f16, fp8, fp4 and quantized integers
- Post-ops fusion of activations and quantization into matmul and convolution
- Graph API fusing attention patterns such as SDPA and GQA
Best for: Deep learning application and framework developers improving performance on supported CPUs and GPUs.
Good for
- Open-source library with binary installs via conda-forge and Intel oneAPI
- Single API covers Intel64/AMD64, AArch64, RISC-V CPUs and Intel GPUs
- Used as acceleration layer by PyTorch, TensorFlow, OpenVINO and ONNX Runtime
Watch out for
- Support for NVIDIA GPU, AMD GPU, PPC64, s390x and RISC-V is experimental
- Designed for framework developers; practitioners are directed to use enabled applications
oneDNN at a glance
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
oneDNN alternatives
Tools and sites you could use instead, ranked by how closely they match.
DyNet provides tools for building dynamic neural networks and deep learning models.
Open source
24 alternatives
The Darknet neural network framework is written in C and CUDA and supports CPU and GPU computation.
Open source
24 alternatives
einops provides readable tensor operations for PyTorch, JAX, TensorFlow, and other frameworks.
Open source
24 alternatives
Deep learning in Rust, with shape checked tensors and neural networks.
Catniff provides JavaScript developers with a Torch-like deep learning framework.
Open source
24 alternativesPopular in AI Infrastructure & MLOps
Well-known listings from the same topic.
Guild AIguild.aiFlow AIflow-ai.comsystemprompt.iosystemprompt.ioCoreWeavecoreweave.comKili Technologykili-technology.comAPI docsdocs.x.ai
oneDNN FAQ
How can oneDNN be installed?
Pre-built packages are available from conda-forge with conda install -c conda-forge onednn and from Intel oneAPI, or it can be built from source.
Which hardware is fully supported versus experimental?
It is optimized for Intel 64/AMD64, AArch64 and Intel Graphics; support for NVIDIA GPU, AMD GPU, PPC64, s390x and RISC-V is experimental.
Which frameworks use oneDNN?
PyTorch, TensorFlow, OpenVINO, ONNX Runtime and other frameworks and applications including llama.cpp, PaddlePaddle and Apache SINGA.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.