If the GPUs can run C++ code, it is likely that they'll be able to run Rust code as well. If there is an LLVM target for it, Rust usually runs on it.
Rust has yet to define a memory model.
I also can't find any details on the CUDA project you mention, it seems counter-intuitive to me. At least with graphics data, it's mostly just large buffers of structs, defined with vertex attrib pointers or whatever. What advantages would a more complex memory model yield?
https://arxiv.org/pdf/1803.04432.pdf
https://people.mpi-sws.org/~viktor/slides/2016-01-debug.pdf
A language memory model is a mathematical specification on how the language semantics map to actual CPUs and the basis how to write lock-free algorithms.
If you bothered to follow the YouTube videos you would found the CUDA projects I mentioned.
Rust currently doesn't have such mathematical model.
Dude, you linked to two hours worth of video. Give people a break.
Just clicking on the links would show the info.
consistency