We also use tools a lot. We're also not very good with mental arithmetic but we can do quite a lot with a simple tool like a piece of paper and a pencil.

Obviously LLM models cannot be compared one to one with our brains, but one thing is clear: despite them being executed by a computer, they are not the computer that is executing them and as such their ability to do operations that are trivial for the computer is not necessarily trivial for the language model.