←back to thread

Show HN: HipScript – Run CUDA in the browser with WebAssembly and WebGPU

(hipscript.lights0123.com)
309 points lights0123 | 5 comments | 07 Jan 25 15:44 UTC | HN request time: 0.206s | source

CUDA is NVIDIA's language for GPU programming, allowing you to mix write CPU and GPU code in C++ in one file. By chaining a few projects that compile CUDA to OpenCL, then Vulkan, then WebGPU, you can experiment with this GPGPU language on any hardware.
1. punnerud ◴[07 Jan 25 20:35 UTC] No.42627113[source]▶
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?

replies(3): >>42627457 #>>42632327 #>>42632538 #
2. mordechai9000 ◴[07 Jan 25 21:06 UTC] No.42627457[source]▶
Distributed large model training with web clients?
replies(1): >>42629054 #
3. sroussey ◴[07 Jan 25 23:36 UTC] No.42629054[source]▶
With 10Tb/s Internet, then sure!
4. machinekob ◴[08 Jan 25 08:50 UTC] No.42632327[source]▶
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.
5. saagarjha ◴[08 Jan 25 09:42 UTC] No.42632538[source]▶
Well the bigger issue is that even natively I doubt there is enough performance to train anything interesting.