slurmstepd: error: Detected 1 oom-kill event(s) in step 1958156.batch cgroup. Some of your processes may have been killed by the cgroup out-of-memory handler.
It seeems the application reises high levels of memory consuption. The loading of all models before skimming is probably the main cause.
The best would be to parallelize in a way that a a model can be started to be skimmed as soon as it is loaded, before waiting the rest to load.