QCDNUM is a very fast QCD evolution program written in FORTRAN77. The releases 17-01/14 and higher come with a C++ interface so that QCDNUM routines can be called from a C++ program.

QCDNUM numerically solves the DGLAP evolution equations on a discrete grid in x and Q2. You can evolve unpolarised parton density functions in NNLO, and polarised pdfs or fragmentation functions in NLO. The jobs example.f and example.cc show how to do this in a few lines of code.

QCDNUM supports evolution in the fixed and variable number schemes. To study the scale uncertainties, the renormalisation scale can be varied with respect to the factorisation scale.

QCDNUM also provides a large toolbox where you can enter your own splitting functions and then solve N-fold coupled DGLAP evolution equations (this is available in the releases 17-01 and higher). Also provided are routines to convolute pdfs with each other, or with user-defined kernels. With these tools you can write your own evolution routine (e.g. QCD-QED) or calculate parton luminosities and structure functions in both the massless and generalised mass schemes.

With the toolbox you can extend the functionality of QCDNUM as is done by the structure function add-on packages ZMSTF (unpolarised zero-mass structure functions) and HQSTF (heavy quark structure functions in the fixed-flavour number scheme).

Also included in the distribution is the MBUTIL collection of FORTRAN routines which should be linked to QCDNUM; apart from this the program is completely stand-alone.

If you use QCDNUM, please refer to

M. Botje, Comput. Phys. Commun. 182 (2011) 490, arXiv:1005.1481, Erratum arXiv:1602.08383 (2016)

The most recent stable version is QCDNUM-18-00/00, released on March 8, 2022. This version is, apart from a few minor fixes, identical to the pre-release QCDNUM-17-01/15 (31-Oct-2019).

For the evolution of fragmentation functions please do not use versions prior to 17-00/07 and 17-01/12 (see erratum).