Ask HN: How are you running AI experiments?
1 comments
(OP here) I'll start: I'm trying to get a nice Stable Diffusion setup going so I can run my own LORAs and enable non-technical designers and artists in my circle to use it.
After realizing my old GPU at home isn't good enough and I should use cloud (duh!) running into struggles around how to set something up that doesn't run all the time (cost is too high) but can turn on/off based on usage or a request that a non-technical user can perform without having to get access to the cloud account. Also what if they forget to turn it off!
This is a great use case for serverless but there's no GPU access from serverless platforms at the moment. Also some issues getting the major vendors to give me a GPU-enabled account!
Would love to hear how others have faced this challenge.
After realizing my old GPU at home isn't good enough and I should use cloud (duh!) running into struggles around how to set something up that doesn't run all the time (cost is too high) but can turn on/off based on usage or a request that a non-technical user can perform without having to get access to the cloud account. Also what if they forget to turn it off!
This is a great use case for serverless but there's no GPU access from serverless platforms at the moment. Also some issues getting the major vendors to give me a GPU-enabled account!
Would love to hear how others have faced this challenge.
What vendors are you using, what roadblocks have you faced, how have you overcome them, what are some up and coming tools other HNers should be trying. And, of course, what you are working on!
There was a post earlier today from a new GPU cloud service but it made me think it would be great to hear from the community what's good right now.