Comment 4 for bug 2084587

This will avoid the issue that you do not have stdhep installed (or even installable on your system).

But you have here at least two others issue:

1)WARNING: All Born diagrams do not factorize the same sum of power(s) of the the perturbed order(s) QCD.
This is potentially dangerous as the real-emission diagrams from aMC@NLO will not be consistent with these virtual contributions.

This warning is quite serious, and is typically followed later in the code by a crash (in version 2.9.x) when computing the actual power of alpha_s for the born. (version 3 is more flexible here).
If this warning is what you expect then 2.9.x is not for you and you need to use 3.x for such kind of NLO computation (but you need to use the framework of double expansion in that case and be clear what type of correction you want to include --3.1 will likely complain that your syntax is ambiguous on that prospect).

Note that you actually do not see such crash yet in 2.9, since you face another crash first (see next point).

2) str : Some tests failed, run cannot continue.

So here, you have an issue with the behaviour of your amplitude in the soft limit:
technical detail in the configuration where the soft limit is not correct:
FKS configuration number is 1
 FKS partons are: i= 5 j= 1
 with PDGs: i= 21 j= 2
for config =1

In all NLO compuation, we first check for a couple of phase-space pont the soft and collinear limit of the amplitude in order to check that substraction of the divergence will be done properly for the process.
For your BSM model, looks like one soft limit of your amplitude is not the compatible with the FKS substraction scheme and therefore the result will not be NLO accurate.

Here I do not know if this is related to a bad handling of the double expansion (like pointed above) or a more serious issue.

Cheers,

Olivier