---
id: onemkl-sycl-blas
version: "2026.1.0"
license: Intel Simplified Software License
license_treatment: unclear
maintenance: active
---

# onemkl-sycl-blas — Intel® oneAPI Math Kernel Library

License: unclear · Maintenance: active · Downloads: 171.8K/mo

## What it is and what it does

Intel oneAPI Math Kernel Library (oneMKL) is Intel's production-grade library for mathematical computing, providing highly optimized routines for linear algebra, Fourier transforms, and related operations. This package exposes C and Data Parallel C++ (DPC++) language interfaces, allowing you to call these routines directly from C, C++, or any language with C bindings. It targets maximum performance on current and future Intel CPUs and GPUs.

The package is intended for developers and researchers who need to optimize compute-intensive mathematical workloads. It requires three runtime dependencies (dpcpp-cpp-rt, intel-opencl-rt, mkl) and manual configuration of the MKLROOT environment variable after installation. The library is actively maintained and carries a proprietary Intel license.

Use it for:

- Optimize matrix multiplication and linear algebra operations in C/C++ applications targeting Intel hardware
- Accelerate scientific computing workloads on Intel GPUs using DPC++ data-parallel programming
- Replace or supplement standard BLAS/LAPACK implementations with Intel's tuned routines for performance-critical code
- Develop high-performance numerical simulations in physics, chemistry, or machine learning on Intel platforms

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides optimized C and Data Parallel C++ (DPC++) interfaces to Intel's oneAPI Math Kernel Library for high-performance linear algebra and mathematical computing on Intel CPUs and GPUs.

Yes, if you are developing C or C++ applications that require high-performance mathematical routines on Intel hardware and are willing to manage the runtime dependencies and MKLROOT configuration. Not recommended for pure Python workflows (use NumPy/SciPy instead) or if you need cross-platform portability beyond Linux and Windows. Verify license compatibility with your project before committing.

## Install

pip install onemkl-sycl-blas
uv add onemkl-sycl-blas
poetry add onemkl-sycl-blas

## Installing onemkl-sycl-blas

Before you install: Medium install friction due to three runtime dependencies (dpcpp-cpp-rt, intel-opencl-rt, mkl) that must be present. Actively maintained with a recent release. Note that MKLROOT environment variable requires manual setup post-installation.

License in practice: Licensed under Intel Simplified Software License (proprietary, not SPDX-identified). License treatment is unclear, so review Intel's terms before integrating into your project to confirm compatibility with your use case.

Quickstart:

# Install
pip install onemkl-sycl-blas

# Import and use (C/DPC++ interfaces available)
import onemkl_sycl_blas
# Call optimized routines via C or DPC++ language bindings

MKLROOT environment variable must be configured manually after installation; dpcpp-cpp-rt, intel-opencl-rt, and mkl runtime dependencies must be installed and available on your system.

Verify before relying:

- Exact Python version compatibility (requires_python is unspecified in metadata)
- Whether DPC++ compiler and Intel GPU drivers are required for GPU acceleration
- Specific performance gains vs. standard NumPy/SciPy on typical workloads
- Whether MKLROOT setup is documented in the package itself or only in external Intel docs

## Package facts

- License: Intel Simplified Software License (unclear)
- Python support: unspecified
- Install friction: medium
- Maintenance: active
- Downloads: 171.8K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

intel mkl blas linear algebra, oneapi math kernel library, dpc++ gpu computing, high-performance matrix operations, intel cpu gpu optimization, oneapi dpc++ interfaces, optimized math routines, intel-hardware, high-performance-computing, c-cpp-bindings

[View on SkillFed](https://skillfed.io/packages/onemkl-sycl-blas) · [View on PyPI](https://pypi.org/project/onemkl-sycl-blas/)