I wonder if this type of work can be applied towards translating kernels between GPU vendors, e.g. CUDA → AMD. Does anyone know if that's possible or whether that kind of problem is AGI-complete?

▲

UncleOxidant 3 hours ago | parent | next [-]

It seems like it could be possible now with a bit of work. I don't think that it would require AGI. Didn't AMD have (or fund) something like this and then decide not to pursue it further recently? It was called HIP. There's also ZLUDA https://www.blopig.com/blog/2024/03/an-open-source-cuda-for-...

	▲	measurablefunc 3 hours ago \| parent [-]
		Very interesting.

▲

jryio 3 hours ago | parent | prev [-]

There's a higher level of abstraction

https://www.modular.com/mojo

	▲	measurablefunc 3 hours ago \| parent [-]
		So if CUDA could be ported to Mojo w/ AI then it would be basically available for any GPU/accelerator vendor. Seems like the right kind of approach towards making CUDA a non-issue.