undefined | Better HN

0 pointsqeternity3mo ago0 comments

PyTorch is only part of it. There is still a huge amount of CUDA that isn’t just wrapped by PyTorch and isn’t easily portable.

0 comments

svara3mo ago

... but not in deep learning or am I missing something important here?

Yes, absolutely in deep learning. Custom fused CUDA kernels everywhere.

Yep. MoE, FlashAttention, or sparse retrieval architectures for example.

j / k navigate · click thread line to collapse

0 pointsqeternity3mo ago0 comments

PyTorch is only part of it. There is still a huge amount of CUDA that isn’t just wrapped by PyTorch and isn’t easily portable.

svara3mo ago

... but not in deep learning or am I missing something important here?

Yes, absolutely in deep learning. Custom fused CUDA kernels everywhere.

Yep. MoE, FlashAttention, or sparse retrieval architectures for example.

j / k navigate · click thread line to collapse