Good morning,
I am a beginner user of the hugging face interface and modes. I have successfully created a code that is able to generate text from a prompt (more or less as Chat-GPT works), and it runs on my personal computer. Generally it takes 3 hours to generate a single response. For example if I give the prompt “describe the difference between nuclear fusion and fission” after 3 hours it gives me the full description of the difference.
At home I have multiple unused pcs that maybe can be used to speed up the process. In particular my question is: is it possible to “distribute the work” on multiple computers connected together via the internet?
Excuse me if I am not precise but it is the third time that I work seriously with this kind of models. If something else is needed for understanding better my question don’t esitate to ask.
Thank you in advance