Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
punnerud
8 months ago
|
parent
|
context
|
favorite
| on:
Show HN: HipScript – Run CUDA in the browser with ...
How performant is the CUDA code in browser compared to standalone program?
Could we have PyTorch / ML training with CUDA through the browser performing ok?
mordechai9000
8 months ago
|
next
[–]
Distributed large model training with web clients?
sroussey
8 months ago
|
parent
|
next
[–]
With 10Tb/s Internet, then sure!
saagarjha
8 months ago
|
prev
|
next
[–]
Well the bigger issue is that even natively I doubt there is enough performance to train anything interesting.
machinekob
8 months ago
|
prev
[–]
WebGPU don't have access to most accelerators in GPU so it would be super slow compared to CUDNN version but pure CUDA would be only x times times faster.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
Could we have PyTorch / ML training with CUDA through the browser performing ok?