Hello!
I was reading about Hugging face serverless inference documentation, there are plenty of warm models readily available for one to experiment with.
Although I wonder, some of these models supports context window ranging from 32k to 128k. Can one truly utilize these models at their highest length?
I’m currently under the impression that there are a certain limit put to it given the resources needed to run an LLM.
Additionally, regarding the ToS:
“Content” refers to any material posted, displayed, or accessed on our Website or Hub, including but not limited to code, data, text, graphics, images, applications, or software you, we, or any third party provide or make available.
So, out of curiosity, would the output you generate from using Inference API falls under “content”?
As in, you are solely responsible for it while owning the content you create?
A clarification on this would be greatly appreciated. Thank you in advance!