I want to talk about WebGPU

Re:I want to talk about WebGPU by sebzim4500 on Wednesday May 03, 2023 @11:09AM (#35803322)

It's not so much that people want to run the models in the browser. They want to be able to write and publish one desktop app that e.g. runs a LLaMA quality model and runs decently across the hundreds of different GPUs that exist on their users' machines.

[ Parent | on HN ]

Re:I want to talk about WebGPU by tikkun on Wednesday May 03, 2023 @11:29AM (#35803580)

So it's more of a CUDA alternative for building desktop apps that are GPU powered?

[ Parent | on HN ]

Re:I want to talk about WebGPU by dm33tri on Wednesday May 03, 2023 @03:20PM (#35806533)

Nowhere near CUDA. Maybe OpenCL and Metal replacement because nobody bothers to support them, so just a fallback option for AMD and ARM chips.

[ Parent | on HN ]

Re:I want to talk about WebGPU by tikkun on Wednesday May 03, 2023 @04:57PM (#35807691)

So then:

If your app needs CUDA, you'd need to write it in CUDA.

If you don't need CUDA, you'd write it for WebGPU instead?

If you benefit from CUDA but don't strictly require it, then you can write it in CUDA with a WebGPU fallback.

Is that right?

[ Parent | on HN ]

Re:I want to talk about WebGPU by dm33tri on Thursday May 04, 2023 @08:00AM (#35814023)

I think you only write WebGPU code if you are using JS or Rust for your hobby project

[ Parent | on HN ]