search

declare-lab / tango

A family of diffusion models for text-to-audio generation.
https://tango2-web.github.io/
Other
991 stars 79 forks source link

Hardware. #41

Open arthurwolf opened 6 months ago

arthurwolf commented 6 months ago

Hello.

Amazing work.

What kind of hardware should I expect to need to be able to run the model?

Thank you.

ZhikangNiu commented 5 months ago

I try to train this model and find if we use flan-t5-large as text encoder. The GPU memory needs >32G. I change the text-encoder as flan-t5-base, it needs 32G

jmanhype commented 5 months ago

if i wanted to run this as a serivce what route would you reccomend? I want to host it somehere and use it as an api