even putting aside how rust ownership semantics map poorly onto gpu programming, ml researchers will never learn rust, this will never ever happen...

▲

pjmlp 3 months ago | parent | next [-]

While I agree in principle, CUDA is more than only AI, as people keep forgetting.

▲

taminka 3 months ago | parent [-]

everyone else who uses cuda isn't going to learn rust either

	▲	pjmlp 3 months ago \| parent [-]
		First Rust needs to have tier 1 support for CUDA, in a way that doesn't feel like yak shaving when coding for CUDA.

▲

int_19h 3 months ago | parent | prev | next [-]

The ML researchers are relying on libraries written by someone else. Today, those libraries are mostly C++, and they would benefit from Rust same as most other C++ codebases.

▲

malcolmgreaves 3 months ago | parent | prev | next [-]

ML reachers don’t write code, they ask ChatGPT to make a horribly inefficient, non-portable notebook that has to be rewritten from scratch :)

	▲	staunton 3 months ago \| parent [-]
		It's made easier by that notebook only having to work just once, to produce some plots for the paper/press release/demo.

▲

saagarjha 3 months ago | parent | prev | next [-]

I don’t think this is true. It seems to me more that nobody has put in a serious effort to make a nice interface build using Rust.

▲

the__alchemist 3 months ago | parent | prev [-]

GPGPU programming != ML.