Accelerate math processing routines, increase application performance, and reduce development time.
For the most current functional and security features, update to the latest version as it becomes available.
Get Intel® oneAPI Math Kernel Library (oneMKL)
This site is protected by reCAPTCHA and the Google <a href="https://policies.google.com/privacy">Privacy Policy</a> and <a href="https://policies.google.com/terms"> Terms of Service</a> apply.
Install with Spack
spack install intel-oneapi-mkl
For more information, refer to Spack documentation.
For the next steps, see the Get Started Guide.
Install with YUM
sudo yum install intel-oneapi-mkl-devel
The following packages are available for installation:
intel-oneapi-mkl-devel: Complete oneMKL package for developmentintel-oneapi-mkl: Complete oneMKL package for runtime onlyintel-oneapi-mkl-classic-devel: oneMKL development package for C/Fortran functionality and Cluster componentsintel-oneapi-mkl-classic-includeintel-oneapi-mkl-classic: oneMKL runtime package for C/Fortran functionality and Cluster componentsintel-oneapi-mkl-cluster-develintel-oneapi-mkl-clusterintel-oneapi-mkl-core-develintel-oneapi-mkl-coreintel-oneapi-mkl-sycl-devel: oneMKL development package for SYCL functionality and GPU supportintel-oneapi-mkl-sycl-includeintel-oneapi-mkl-sycl: oneMKL runtime package for SYCL functionality and GPU supportintel-oneapi-mkl-sycl-blas: oneMKL runtime package for SYCL functionality and GPU support for BLAS onlyintel-oneapi-mkl-sycl-data-fitting: oneMKL runtime package for SYCL functionality and GPU support for Data Fitting only (experimental library)intel-oneapi-mkl-sycl-dft: oneMKL runtime package for SYCL functionality and GPU support for the Fast Fourier transforms onlyintel-oneapi-mkl-sycl-lapack: oneMKL runtime package for SYCL functionality and GPU support for LAPACK onlyintel-oneapi-mkl-sycl-rng: oneMKL runtime package for SYCL functionality and GPU support for Random Number Generators onlyintel-oneapi-mkl-sycl-sparse: oneMKL runtime package for SYCL functionality and GPU support for Sparse BLAS onlyintel-oneapi-mkl-sycl-stats: oneMKL runtime package for SYCL functionality and GPU support for Summary Statistics onlyintel-oneapi-mkl-sycl-vm: oneMKL runtime package for SYCL functionality and GPU support for Vector Math only- For distributed SYCL-based applications:
intel-oneapi-mkl-sycl-distributed-dft: oneMKL runtime package for SYCL functionality and GPU support for distributed Discrete Fourier Transform only (experimental library)intel-oneapi-mkl-sycl-distributed-dft-devel: oneMKL development package for SYCL functionality and GPU support for distributed Discrete Fourier Transform only (experimental library)
For the next steps, see the Get Started Guide.
- Install Miniforge.
- Create and activate a new conda environment, replacing
<your-env-name>with your preferred name for the environment:conda create -n <your-env-name> conda activate <your-env-name>
conda install -c https://software.repos.intel.com/python/conda/ -c conda-forge <package-name>
The following packages are available for installation:
- For running applications that require oneMKL:
mklincludes runtime only
- For developing or compiling applications:
mkl-develincludes libraries, headers, and tools for dynamic linking- If needed, add
mkl-includeif your development workflow manages the libraries separately
- For static linking:
mkl-staticto statically link oneMKL, creating self-contained binaries
- For running Data Parallel C++ (DPC++) applications that require oneMKL:
mkl-dpcppprovides the runtime support for oneMKL with DPC++
- For developing and compiling DPC++ applications:
mkl-devel-dpcppincludes the development tools and headers for oneMKL with DPC++
- For using domain-specific libraries in SYCL-based applications:
onemkl-sycl-blasprovides Basic Linear Algebra Subprograms (BLAS) routinesonemkl-sycl-lapackprovies Linear Algebra Package (LAPACK) routines for more advanced linear algebra computationsonemkl-sycl-dftprovides Discrete Fourier Transform functionalityonemkl-sycl-sparseprovides sparse matrix operationsonemkl-sycl-vmprovides vector math (VM) operations, which optimize common mathematical functions applied to vectorsonemkl-sycl-datafittingprovides functionality for data fitting operations
For the next steps, see the Get Started Guide.
- Install Miniforge.
- Create and activate a new conda environment, replacing
<your-env-name>with your preferred name for the environment:conda create -n <your-env-name> conda activate <your-env-name>
conda install -c https://software.repos.intel.com/python/conda/ -c conda-forge <package-name>
The following packages are available for installation:
- For running applications that require oneMKL:
mklincludes runtime only
- For developing or compiling applications:
mkl-develincludes libraries, headers, and tools for dynamic linking- If needed, add
mkl-includeif your development workflow manages the libraries separately
- For static linking:
mkl-staticto statically link oneMKL, creating self-contained binaries
- For single-device SYCL-based applications:
- For running SYCL-based applications that require oneMKL:
mkl-dpcppprovides the runtime support for oneMKL with SYCL
- For developing and compiling SYCL-based applications:
mkl-devel-dpcppincludes the development tools and headers for oneMKL with SYCL
- For using domain-specific libraries when running SYCL-based applications:
onemkl-sycl-blasprovides Basic Linear Algebra Subprograms (BLAS) routinesonemkl-sycl-lapackprovides Linear Algebra Package (LAPACK) routines for more advanced linear algebra computationsonemkl-sycl-dftprovides Discrete Fourier Transform functionalityonemkl-sycl-sparseprovides sparse matrix operationsonemkl-sycl-vmprovides vector math (VM) operations, which optimize common mathematical functions applied to vectors- (Experimental feature)
onemkl-sycl-datafittingprovides functionality for data fitting operations
- For running SYCL-based applications that require oneMKL:
- For distributed SYCL-based applications:
- For using domain-specific libraries when running SYCL-based applications:
- (Experimental feature)
onemkl-sycl-distributed-dftprovides runtime support for distributed Discrete Fourier Transform functionality.
- (Experimental feature)
- For using domain-specific libraries when running SYCL-based applications:
For the next steps, see the Get Started Guide.
Install with NuGet
You can install NuGet packages for oneMKL via Microsoft* Visual Studio or command line interface. For more information, refer to the NuGet documentation.
The following packages are available for installation
- Development packages contain the development libraries and headers, enabling you to compile and link your applications with the oneMKL libraries:
intelmkl.devel.win-x64intelmkl.devel.win-x86 - Static packages contain the static libraries, allowing you to link oneMKL libraries statically into your application. This means your application will not depend on external dynamic libraries at runtime, simplifying distribution:
intelmkl.static.win-x64intelmkl.static.win-x86
Additionally, oneMKL cluster components development and static packages are available: intelmkl.devel.cluster.win-x64 intelmkl.static.cluster.win-x64
For the next steps, see the Get Started Guide.
Create and activate a virtual environment, replacing <your-env-name> with your preferred name for the environment:
python3.10 -m venv <your-env-name>
source <your-env-name>/bin/activate
sudo pip install <package-name>
The following packages are available for installation:
- For running applications that require oneMKL:
mklincludes runtime only
- For developing or compiling applications:
mkl-develincludes libraries, headers, and tools for dynamic linking- if needed, add
mkl-includeif your development workflow manages the libraries separately
- For static linking:
mkl-staticto statically link oneMKL, creating self-contained binaries
- For running Data Parallel C++ (DPC++) applications that require oneMKL:
mkl-dpcppprovides the runtime support for oneMKL with DPC++
- For developing and compiling DPC++ applications:
mkl-devel-dpcppincludes the development tools and headers for oneMKL with DPC++
- For using domain-specific libraries in SYCL-based applications:
onemkl-sycl-blasprovides Basic Linear Algebra Subprograms (BLAS) routinesonemkl-sycl-lapackprovies Linear Algebra Package (LAPACK) routines for more advanced linear algebra computationsonemkl-sycl-dftprovides Discrete Fourier Transform functionalityonemkl-sycl-sparseprovides sparse matrix operationsonemkl-sycl-vmprovides vector math (VM) operations, which optimize common mathematical functions applied to vectorsonemkl-sycl-datafittingprovides functionality for data fitting operations
For the next steps, see the Get Started Guide.
- To add APT repository access, install the prerequisites:
sudo apt update sudo apt install -y gpg-agent wget - Set up the repository. To do this, download the key to the system keyring:
wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | sudo tee /usr/share/keyrings/oneapi-archive-keyring.gpg > /dev/null - Add the signed entry to APT sources and configure the APT client to use the Intel repository:
echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt.repos.intel.com/oneapi all main" | sudo tee /etc/apt/sources.list.d/oneAPI.list - Update the packages list and repository index.
sudo apt update
For running applications that require oneMKL:
sudo apt install intel-oneapi-mkl
For developing and compiling oneMKL applications:
sudo apt install intel-oneapi-mkl-devel
The following packages are available for installation:
intel-oneapi-mkl-devel: Complete oneMKL package for developmentintel-oneapi-mkl: Complete oneMKL package for runtime onlyintel-oneapi-mkl-classic-devel: oneMKL development package for C/Fortran functionality and Cluster componentsintel-oneapi-mkl-classic-includeintel-oneapi-mkl-classic: oneMKL runtime package for C/Fortran functionality and Cluster componentsintel-oneapi-mkl-cluster-develintel-oneapi-mkl-clusterintel-oneapi-mkl-core-develintel-oneapi-mkl-coreintel-oneapi-mkl-sycl-devel: oneMKL development package for SYCL functionality and GPU supportintel-oneapi-mkl-sycl-includeintel-oneapi-mkl-sycl: oneMKL runtime package for SYCL functionality and GPU supportintel-oneapi-mkl-sycl-blas: oneMKL runtime package for SYCL functionality and GPU support for BLAS onlyintel-oneapi-mkl-sycl-data-fitting: oneMKL runtime package for SYCL functionality and GPU support for Data Fitting only (experimental library)intel-oneapi-mkl-sycl-dft: oneMKL runtime package for SYCL functionality and GPU support for the Fast Fourier transforms onlyintel-oneapi-mkl-sycl-lapack: oneMKL runtime package for SYCL functionality and GPU support for LAPACK onlyintel-oneapi-mkl-sycl-rng: oneMKL runtime package for SYCL functionality and GPU support for Random Number Generators onlyintel-oneapi-mkl-sycl-sparse: oneMKL runtime package for SYCL functionality and GPU support for Sparse BLAS onlyintel-oneapi-mkl-sycl-stats: oneMKL runtime package for SYCL functionality and GPU support for Summary Statistics onlyintel-oneapi-mkl-sycl-vm: oneMKL runtime package for SYCL functionality and GPU support for Vector Math only- For distributed SYCL-based applications:
intel-oneapi-mkl-sycl-distributed-dft: oneMKL runtime package for SYCL functionality and GPU support for distributed Discrete Fourier Transform only (experimental library)intel-oneapi-mkl-sycl-distributed-dft-devel: oneMKL development package for SYCL functionality and GPU support for distributed Discrete Fourier Transform only (experimental library)
For the next steps, see the Get Started Guide.
Install with Cloudera
- In the Cloudera Manager Admin Console, access the Parcels page by clicking the Parcels indicator in the left navigation bar.
- At the Parcels page, click the Parcel Repositories & Network Settings button.
- In the Remote Parcel Repository URLs list, click the plus symbol to open an additional row. Enter the path to oneMKL Parcel repository:
http://parcels.repos.intel.com/mkl/latest. Click the Save & Verify configuration button. - Click the Check for New Parcels button. In the Location selector, click Available Remotely. The latest oneMKL parcel should be available for download.
- Click the Download button for the oneMKL parcel. By downloading the package, you agree with the terms and conditions stated in the End-User License Agreement (EULA).
- When download is completed, click the Distribute button to distribute the parcel on all cluster nodes.
- When distribution is completed, click the Activate button to activate the parcel on all cluster nodes.
For additional information about Clouder parcels, refer to Parcels documentation.
For the next steps, see the Get Started Guide.
Install with DNF
sudo dnf install intel-oneapi-mkl-devel
For the next steps, see the Get Started Guide.
Create and activate a virtual environment, replacing <your-env-name> with your preferred name for the environment:
python3.10 -m venv <your-env-name>
source <your-env-name>/bin/activate
sudo pip install <package-name>
The following packages are available for installation:
- For running applications that require oneMKL:
mklincludes runtime only
- For developing or compiling applications:
mkl-develincludes libraries, headers, and tools for dynamic linking- If needed, add
mkl-includeif your development workflow manages the libraries separately
- For static linking:
mkl-staticto statically link oneMKL, creating self-contained binaries
- For single-device SYCL-based applications:
- For running SYCL-based applications that require oneMKL:
mkl-dpcppprovides the runtime support for oneMKL with SYCL
- For developing and compiling SYCL-based applications:
mkl-devel-dpcppincludes the development tools and headers for oneMKL with SYCL
- For using domain-specific libraries when running SYCL-based applications:
onemkl-sycl-blasprovides Basic Linear Algebra Subprograms (BLAS) routinesonemkl-sycl-lapackprovides Linear Algebra Package (LAPACK) routines for more advanced linear algebra computationsonemkl-sycl-dftprovides Discrete Fourier Transform functionalityonemkl-sycl-sparseprovides sparse matrix operationsonemkl-sycl-vmprovides vector math (VM) operations, which optimize common mathematical functions applied to vectors- (Experimental feature)
onemkl-sycl-datafittingprovides functionality for data fitting operations
- For running SYCL-based applications that require oneMKL:
- For distributed SYCL-based applications:
- For using domain-specific libraries when running SYCL-based applications:
- (Experimental feature)
onemkl-sycl-distributed-dftprovides runtime support for distributed Discrete Fourier Transform functionality.
- (Experimental feature)
- For using domain-specific libraries when running SYCL-based applications:
For the next steps, see the Get Started Guide.
Install with Zypper
sudo zypper install intel-oneapi-mkl-devel
For the next steps, see the Get Started Guide.
By downloading, you agree to our Privacy and Terms of use
Thank You
Your Download should start immediately.
Failed to submit your form.
Due to a technical difficulty, we were unable to submit the form. Please try again after a few minutes. We apologize for the inconvenience.
Installation Instructions
The initial download includes the installer application files only. The installer will acquire the component during the installation process.
Step 1: Select the .exe file to launch the GUI installer.
Step 2: Follow the instructions in the installer.
Step 3: Explore the Get Started Guide.
Command Line Download
Command Line Installation Parameters
wget https://registrationcenter-download.intel.com/akdlm/IRC_NAS/47c7d946-fca1-441a-b0df-b094e3f045ea/intel-onemkl-2025.2.0.629.sh
sudo sh ./intel-onemkl-2025.2.0.629.sh
Command Line Download
Command Line Installation Parameters
wget https://registrationcenter-download.intel.com/akdlm/IRC_NAS/47c7d946-fca1-441a-b0df-b094e3f045ea/intel-onemkl-2025.2.0.629_offline.sh
sudo sh ./intel-onemkl-2025.2.0.629_offline.sh
Installation Instructions
Step 1: From the console, locate the downloaded install file.
Step 2: Use $ sudo sh ./<installer>.sh to launch the GUI Installer as the root.
Optionally, use $ sh ./<installer>.sh to launch the GUI Installer as the current user.
Step 3: Follow the instructions in the installer.
Step 4: Explore the Get Started Guide.
Create the DNF repository file in the /temp directory as a normal user.
tee > /tmp/oneAPI.repo << EOF
[oneAPI]
name=Intel® oneAPI repository
baseurl=https://yum.repos.intel.com/oneapi
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://yum.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB
EOF
Move the newly created oneAPI.repo file to the YUM configuration directory.
sudo mv /tmp/oneAPI.repo /etc/yum.repos.d
Prerequisites for First-Time Users
- Create the YUM repository file in the /temp directory as a normal user.
tee > /tmp/oneAPI.repo << EOF [oneAPI] name=Intel® oneAPI repository baseurl=https://yum.repos.intel.com/oneapi enabled=1 gpgcheck=1 repo_gpgcheck=1 gpgkey=https://yum.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB EOF
- Move the newly created oneAPI.repo file to the YUM configuration directory /etc/yum.repos.d.
sudo mv /tmp/oneAPI.repo /etc/yum.repos.d
Prerequisites for First-Time Users
Add the Intel oneAPI repository public key using the following command:
sudo zypper addrepo https://yum.repos.intel.com/oneapi oneAPI
Additional Resources
System Requirements
Complete Installation Guide
Release Notes
Intel Simplified Software License
Support
Start-up support is available if there is an issue with the tool selector functionality.