R for Mac OS X

Development Tools and Libraries

This directory contains tools and libraries that are part of the base distribution of R for Mac OS X.

Note: CRAN does not have Mac OS X systems and cannot check these binaries for viruses. Altough we take precautions when assembling binaries, please use the normal precautions with downloaded executables.

Important note: R 3.5.0 El Capitan binaries are using Clang 6.0.0 and GNU Fortran 6.1 to provide OpenMP parallelization support and C++17 standard features. If you want to compile R packages from sources, please download GNU Fortran binary from the official GNU Fortran Binaries page - in particular OS X 10.11 gfortran 6.1. Alternatively, we are providing a copy here as well as Clang 6.0.0 binaries for OS X 10.11 and higher - see below for the download links. You can also try to use clang from Xcode, but it will be missing required features so your mileage may vary and it is not recommended.


clang-6.0.0.pkg (OS X 10.11+, signed, 64-bit)
MD5-hash: c29700c4e7b2914073ef7e741eb105bc
(ca. 418Mb)
Clang 6.0.0 for OS X 10.11 and higher, static build for x86_64, signed package, installs into /usr/local/clang6. To be used with El Capitan builds of R (typically R 3.4.0 and higher).
gfortran-6.1.pkg (OS X 10.11+, signed, 64-bit)
MD5-hash: 201026216e8b373d9cd2efc0cc474bb8
(ca. 73Mb)
GNU Fortran 6.1 for OS X 10.11 and higher - a copy from GFortranBinaries pages for x86_64, signed package, installs into /usr/local/gfortran (identical content, re-packaged to a flat Installer package and signed). To be used with El Capitan builds of R.

The following binaries are obsolete and only provided for historical reasons
gfortran-4.2.3.pkg (OS X 10.5+, signed, 64-bit driver)
MD5-hash: 8783f803038abe6487a362ad5b8995ea
(ca. 27MB)
gfortran-4.2.3.dmg (OS X 10.4, 32-bit driver)
MD5-hash: 9551fc46f55537dd1db581154daf27ef
(ca. 27MB)
Universal GNU Fortran 4.2.3 for Mac OS X 10.4 and higher. It is necessary in order to build R packages from sources that contain Fortran code.

Unlike many other builds, this is a fully universal build of GNU Fortran that uses Apple's driver and supports all target architectures (i386, ppc, x86_64 and ppc64). As such it fully supports compilation into fat files like gfortran -arch i386 -arch ppc -arch x86_64 -arch ppc64 t.f -o t on both Intel Macs and PowerPC Macs (32- and 64-bit). Dependent libraries are fat as well, avoiding problems known from other Fortran builds (such as those from HPC). It installs in /usr/local and comes with an uninstall-script.

tcltk-8.5.5-x11.pkg (OS X 10.5+, signed)
MD5-hash: e7c406d91762ffdc4539b23c5b5a3ab4
(ca. 9MB)
tcltk-8.5.5-x11.dmg (OS X 10.4)
MD5-hash: c32dda1b9f2c2776a02cec4e03befc76
(ca. 9MB)
Universal build of Tcl/Tk 8.5.5 for X11 (32-bit and 64-bit). This library is necessary in order to use the tcltk R package (for R 2.8.0 - 2.15.3 only!). It installs in /usr/local. Requires Mac OS X 10.4 (Tiger) or higher for 32-bit R and Mac OS X 10.5 (Leopard) or higher for 64-bit R.

NOTE: R 3.0.0 and higher comes bundled with Tck/Tk 8.6.0 so you do not need this package

For other (optional) 3rd party libraries for development see http://R.research.att.com/libs/. The devpack has been superseded by those libraries. For R you may want to download and install libpng, libjpeg, readline, freetype, fontconfig, pixman and cairo.

Source code for all 3rd party libraries can be found at http://R.research.att.com/src/


old Previous versions of tools as supplied with legacy R versions.

You may also want to read the R FAQ and R for Mac OS X FAQ. For discussion of Mac-related topics and reporting Mac-specific bugs, please use the R-SIG-Mac mailing list.

Information, tools and most recent daily builds of the R GUI, R-patched and R-devel can be found at http://R.research.att.com/. Please visit that page especially during beta stages to help us test the Mac OS X binaries before final release! The page also contains links to experimental builds as such 64-bit R for OS X.

Link to corresponding sources: http://r.research.att.com/src/

Last modified: 2018/04/30, by Simon Urbanek