Are you training your own model? If so, what makes your model different?

The difference in my experience comes from the data-set. If you have an unusual and proprietary dataset, then off-the-shelf models are only a starting point.

Does fine tuning count? If so there isn't a model for the input I'm working with that I know of

I train LoRa Models or finetune foundation models for the task at hand. I've not encountered the need to train a foundation model, yet.