HF Inference Endpoints Difference between Max Input Length per Query and Max Token Length per Query

Inference Endpoints on the Hub
1

Can anyone explain the difference?

2

Found it – these are explained in more detail in the TGI documentation here