top | item 39413652 Interactive Coloring with SD ControlNet 11 points| matroid | 2 years ago |vrroom.github.io | reply 2 comments order hn newest [+] [-] mlsu|2 years ago|reply Why spend almost $1,000 to rent an A10 GPU? You can purchase a used 3090 outright for that much. [+] [-] matroid|2 years ago|reply Two reasons (may not be well thought out or wrong):* Flash Attention, an efficient attention module which significantly speeds up training, only works on Ampere GPUs [1]* Even if I bought a 3090, I would have to get a computer to go with it, along with a PSU and some cooling. Don't know where to start with that.[1] https://github.com/Dao-AILab/flash-attention/issues/190
[+] [-] mlsu|2 years ago|reply Why spend almost $1,000 to rent an A10 GPU? You can purchase a used 3090 outright for that much. [+] [-] matroid|2 years ago|reply Two reasons (may not be well thought out or wrong):* Flash Attention, an efficient attention module which significantly speeds up training, only works on Ampere GPUs [1]* Even if I bought a 3090, I would have to get a computer to go with it, along with a PSU and some cooling. Don't know where to start with that.[1] https://github.com/Dao-AILab/flash-attention/issues/190
[+] [-] matroid|2 years ago|reply Two reasons (may not be well thought out or wrong):* Flash Attention, an efficient attention module which significantly speeds up training, only works on Ampere GPUs [1]* Even if I bought a 3090, I would have to get a computer to go with it, along with a PSU and some cooling. Don't know where to start with that.[1] https://github.com/Dao-AILab/flash-attention/issues/190
[+] [-] mlsu|2 years ago|reply
[+] [-] matroid|2 years ago|reply
* Flash Attention, an efficient attention module which significantly speeds up training, only works on Ampere GPUs [1]
* Even if I bought a 3090, I would have to get a computer to go with it, along with a PSU and some cooling. Don't know where to start with that.
[1] https://github.com/Dao-AILab/flash-attention/issues/190