Thanks a lot for the work. I just followed all the below mentioned steps,
and it's working without any problem. I just finished running example
problems successfully.
Great! I'm gald your got it all working.
Can you please help me with two below mentioned points:
1.) We don't have to recompile the LAMMPS every-time we add a new model
and/or driver, right?
Correct, you can add KIM models at anytime and there is no need to recompile lammps.
2.) Running on a virtual or local machine is okay for time being but
ultimately I've to run the kim models on our university cluster where I
can't use sudo command. What would you suggest if I want to use kim models
with the jobs on my university cluster in the given scenerio.
You can do it very much like you did on the VM. There was no use of sudo in that case. The only complication may be some fiddling with the compiles that are used to make sure that the kim-api and lammps are built using the same suite of compilers (c, c++, and forgran).
Cheers,
Ryan