Intel® Fortran Compiler
Build applications that can scale for the future with optimized code designed for Intel® architecture.
A Tradition of Trusted Application Performance
The Intel® Fortran Compiler is built on a long history of generating optimized code that supports industry standards while taking advantage of built-in technology for Intel® Xeon® Scalable processors, Intel® Xeon® processors, Intel® Core™ Ultra processors, and Intel® Core™ processors.
Staying aligned with Intel's evolving and diverse architectures, the compiler now supports GPUs.
Download as Part of the Toolkit or the Essentials
The Intel Fortran Compiler is included in the Intel® oneAPI HPC Toolkit and in Intel® Fortran Essentials. Get the toolkit or essentials bundle to analyze, optimize, and deliver applications that scale.
Download the Stand-Alone Version
You can download binaries from Intel or choose your preferred repository.
Features
Cross-Architecture Compilation
- Improves development productivity by targeting CPUs and GPUs through single-source code while permitting custom tuning
- Fully supports broad Fortran language standards up to and including 2018, plus select Fortran 2023 language features
- Incorporates industry standards support for OpenMP* 4.5, and near-complete OpenMP 5.0, 5.1, and 5.2 for CPU and GPU offload, and select new features of OpenMP 6.0
- Uses LLVM* sanitizers to catch bugs and issues early in the development cycle, enhancing the reliability, security, and maintainability of Fortran and OpenMP code on both CPU and GPU platforms
- Uses well-proven LLVM compiler technology and Intel's history of compiler leadership
Built-In Performance
- Takes advantage of multicore, Single Instruction Multiple Data (SIMD) vectorization, and multiprocessor systems with OpenMP, automatic parallelism, coarrays, and DO CONCURRENT for GPU offload
- Optimizes code with an automatic processor dispatch feature
Development Environment Flexibility
Use the compiler in command line or in a supported IDE:
- Microsoft Visual Studio* (Windows only)
- Eclipse* C/C++ Development Tooling (CDT, Linux only)
Benchmark
Code Samples
Fortran and OpenMP* Code Samples for CPU
This code sample demonstrates how to build and run a serial implementation of a Monte Carlo program, and then convert it to run in parallel using Fortran coarrays, providing hands-on experience with the Fortran coarray feature for parallel programming.
The OpenMP Primes code sample demonstrates how to use the OpenMP API with Intel Fortran Compiler to find prime numbers, including the count of specific types of primes. This is done by using two OpenMP directives to enhance code performance, providing practical insight into building and running Fortran OpenMP applications.
Fortran and OpenMP Offload
The Matrix Multiply code sample offers a guided process to adapt a basic Fortran program for offloading computations to an Intel GPU using OpenMP directives with the Intel Fortran Compiler. It focuses on a simple matrix multiplication algorithm to demonstrate the use of OpenMP directives in Fortran for GPU offloading.
OpenMP Offload Fortran Tutorials
This repository contains a Jupyter* Notebook created for illustrating OpenMP offload basics, focusing on the process of offloading intricate calculations to a GPU using OpenMP with the Intel Fortran Compiler, providing comprehensive insights into GPU offloading with OpenMP.
View All Fortran Code Samples in oneAPI Code Samples Catalog
Training
Fortran and OpenMP Offload
The Case for OpenMP Target Offloading [1:06:07]
Learn how industry-standard OpenMP provides a rich set of directives to address the limitations of Fortran, a mature and modern programming language that, nonetheless, evolves slowly.
Offload Fortran Workloads to New Intel GPUs Using OpenMP [2:41]
Explore the offloading capabilities of Intel Fortran Compiler using GRILLIX/PARALLAX, a plasma turbulence and transport simulator.
Accelerator Offload
Find out how to use Fortran DO CONCURRENT for accelerator offload.
Solve the Top Three Programming Challenges with Fortran and OpenMP*
Discover how Fortran and OpenMP solve the three main heterogeneous computing challenges: accelerator offload, disjoint memory management, and API calls.
Accelerate Lower-Upper (LU) Factorization Using Fortran, Intel® oneAPI Math Kernel Library, and OpenMP
Find out how to offload linear algebra computations (specifically, LU factorization) to an accelerator using development tools from Intel.
Upgrade from Intel Fortran Compiler Classic
Porting Guide for Intel Fortran Compiler
These guidelines help the Intel Fortran Compiler Classic users upgrade to Intel Fortran Compiler.
What Customers are Saying
"The INTES FEM solver PERMAS (PERMAS is among the most advanced finite element software systems worldwide) with the new LLVM-based Intel Fortran Compiler, we decided to provide PERMAS also with the Intel Fortran Compiler compiled version as a commercial product. The most recent PERMAS application compiled with latest Intel Fortran Compiler version shows at par performance in comparison to the classical most recent Intel Fortran Compiler Classic version."
"Intel Fortran Compiler outperforms GNU* Fortran (GFortran) by more than 40 percent when testing various scenarios on the complex Stokes Inversion based on Response functions (SIR) numerical code on Ubuntu Linux*. Numerical results were identical between the two compilers, and saved me precious time and effort."
— Carlos Quintero Noda, doctor of astrophysics; International Top Young Fellow, Japan Aerospace Exploration Agency (JAXA)
"I am using Intel Fortran Compiler to develop an automated forest cover identification system from digital aerial images forming stereo pairs. This system uses advanced texture and shape analysis to identify and classify the forest cover (species, density, and height) in order to produce forest inventory maps. Stereo matching is used to produce 3D digital canopy models that are subsequently analyzed by the texture and shape classification program. The advantages of Intel Fortran Compiler are:
- Integration within Microsoft Visual Studio*, thus providing a complete, integrated development environment including a Fortran-aware debugger
- State-of-the-art optimizations for the latest Intel processors (such as Intel® Advanced Vector Extensions instructions) and automatic parallelization
- QuickWin library to rapidly build a Windows interface (the image interpretation system uses the QuickWin library)
— Jean Vezina, software developer and forest engineer
Specifications
Intel Fortran Compiler
Processors:
- Intel Xeon Scalable processors
- Intel Xeon processor family
- Intel® Core™ Ultra processors
- Intel Core processors
- Intel Atom® processors
GPUs:
- Intel® UHD Graphics for 11th generation Intel processors or newer
- Intel® Iris® Xᵉ graphics or newer
- Intel® Arc™ graphics
- Intel® Data Center GPU Flex Series
- Intel® Data Center GPU Max Series
Host and target operating systems:
- Windows
- Linux
Get Help
Your success is our success. Access this support resource when you need assistance.
For additional help, see our general oneAPI Support.
Related Product
Intel® oneAPI DPC++/C++ Compiler
This standards-based, cross-architecture compiler builds high-performance applications by generating optimized code for Intel Xeon Scalable processors, Intel Core processors, Intel Atom processors, and supported GPUs and other accelerators. It allows you to:
- Support open standards of C, C++, SYCL*, and OpenMP
- Generate optimized binary host code and accelerator code
- Work with a familiar set of tools on your preferred platform: Windows, Linux
Stay In the Know on All Things CODE
Sign up to receive the latest tech articles, tutorials, dev tools, training opportunities, product updates, and more, hand-curated to help you optimize your code, no matter where you are in your developer journey. Take a chance and subscribe. You can change your mind at any time.