Comment by dannyw

It is not a trivial challenge setting up model serving infra for ~1T or larger models, especially in a high reliability environment (e.g. your team is using it for work, or you're using it to power production apps). Sure, there are third party providers, although the quality and reliability of their inference varies.