Multi dimensional training

I am new to xgboost. I want to know whether it is possible to train a model that is multi-dimensional (several input parameters but only one output) using xgboost?
If yes, can you please share some examples?
Also, is there a rule of thumb for number of data in training set based on number of inputs?
For instance, I have a FEM model with 10 inputs and one output and I want to train mode ML model, what should be the size of the training set for good accuracy? 100, 1000, …?
Any help is greatly appreciated.