Build and Optimize Applications across CPUs and GPUs

Intel® oneAPI Toolkit is a core set of tools and libraries for developing optimized software that scales across modern systems.

It supports development needs spanning client and edge applications through to large-scale, distributed systems.

The toolkit provides standards-based compilers, domain-specific performance libraries with drop-in acceleration, and profiling, design, and debugging tools. These capabilities help developers build, optimize, and deploy applications efficiently.

It features an industry-leading C++ compiler with support for SYCL*, among other standards-based programming models.

What’s Included

Compilers

Intel® oneAPI DPC++/C++ Compiler

Compile and optimize C++ and SYCL code for CPU and GPU target architectures.

 

Intel® Fortran Compiler

Modern Fortran compiler for performance-critical applications on CPU and GPU.

Performance Libraries

Intel® oneAPI Math Kernel Library

Accelerate math processing routines, including matrix algebra, fast Fourier transforms (FFT), and vector math.

 

Intel® oneAPI Deep Neural Network Library (oneDNN)

Develop fast neural networks on Intel CPUs and GPUs with performance-optimized building blocks.

 

Intel® MPI Library

A high-performance implementation of the MPI standard for building and scaling distributed applications across nodes and clusters.

 

Intel® Integrated Performance Primitives

Speed up performance of imaging, signal processing, data compression, cryptography, and more.

 

Intel® Cryptography Primitives Library

Secure, fast, lightweight building blocks for cryptography optimized for Intel CPUs.

 

Intel® oneAPI Threading Building Blocks

Simplify parallelism with this advanced threading and memory-management template library.

 

Intel® oneAPI DPC++ Library

Speed up data parallel workloads with these key productivity algorithms and functions.

 

Intel® oneAPI Collective Communications Library

Implement optimized communication patterns to distribute deep learning model training across multiple nodes.

Analysis and Debug Tools

Intel® VTune™ Profiler

Find and optimize performance bottlenecks across CPU and GPU systems.

 

Intel® Distribution for GDB*

Enable deep, system-wide debug of SYCL, C, C++, and Fortran code.

Add-ons

Some oneAPI tools and libraries are available as separate downloads and can be added as needed. These components integrate with the oneAPI toolchain but are not included in the default Intel oneAPI Toolkit installers.

 

Intel® Advisor

Design code for efficient vectorization, threading, and offloading to accelerators.

 

Intel® SHMEM

Implement Partitioned Global Address Space (PGAS) programming for host-initiated and device-initiated operations.

 

 Intel® oneAPI Data Analytics Library (oneDAL)

Boost machine learning and data analytics performance.

Specifications

CPUs:

  • Intel® Xeon® Scalable processors
  • Intel® Xeon® processors
  • Intel® Core™ Ultra processors
  • Intel® Core™ processors
  • Intel Atom® processors
  • Other processors compatible with Intel® 64 architecture

 

GPUs:

  • Intel® UHD Graphics for 11th generation Intel processors or newer
  • Intel® Iris® Xe Graphics
  • Intel® Arc™ GPUs
  • Intel® Server GPU
  • Intel® Data Center GPU Flex Series
  • Intel® Data Center GPU Max Series

Operating systems:

  • Linux*
  • Windows*

Languages:

  • C, C++, C++ with SYCL
  • Fortran

Distributed environments:

  • MPI

OpenFabrics Interface* (OFI) framework implementation supporting the following:

  • InfiniBand*
  • iWARP, RDMA over Converged Ethernet (RoCE)
  • Amazon Web Services Elastic Fabric Adapter (AWS EFA)*
  • Cornelis Networks*
  • Ethernet, IP over InfiniBand (IPoIB), IP over Intel OPA

 

Development environments:

  • Microsoft Visual Studio*, Microsoft Visual Studio Code*
  • Eclipse*