⬢github C++ · 64 ★ +51 since we first saw it · pushed 1 d ago · Apache-2.0
ymcrcat/rgpu
Keep Python on your laptop. Run PyTorch operations and hold tensors on a remote GPU, including from a Mac with no CUDA installation.
Rgpu lets your Python/PyTorch code stay on your laptop while tensors and GPU operations execute on a remote NVIDIA machine. It offers a custom 'rgpu' PyTorch device that sends ops over TCP via SSH tunneling, plus a CUDA shim (libcuda, cuBLAS, cuDNN) so unmodified Linux CUDA binaries can also run against the remote GPU — useful for Macs without CUDA.
Why now: It was featured in a Show HN post presenting the PyTorch remote-device demo, drawing attention to using remote GPUs from CUDA-less machines like Macs.
Who it is for: Developers and ML practitioners who want to run PyTorch or CUDA workloads on a remote NVIDIA GPU from a local machine (e.g., a Mac) without local CUDA.
apple-siliconcudadeep-learninggpugpu-computingmachine-learningmlpytorchremote-gpu
Stars over our 270 snapshots: 13 to 64, since 2 d ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/ymcrcat/rgpu