Running Open Source LLM's Like LLama2 on GCP

Hi, does somebody has some experience running some of this Open Source Models in principal Cloud providers such as Azure, GCP or AWS?

I know that they provide some of these models in a catalog, but I’m wondering, what would be the more general WorkFLow for example to run Llama 2 in Google Cloud for example.

I think there is a big gap in complexity between running one of these models locally or with Google Collab and loading a model into a VM an making a connection.

I have found several guides, using for example runpud.io, but not too many with GCP. I will really appreciate some guidance about this topic.