CubeCL: GPU Kernels in Rust for CUDA, ROCm, and WGPU

Re:CubeCL: GPU Kernels in Rust for CUDA, ROCm, and WGPU by gitroom on Thursday April 24, 2025 @04:09AM (#43780320)

Gotta say, the constant dance between all these GPU frameworks kinda wears me out sometimes - always chasing that better build, you know?

[ Parent | on HN ]

Re:CubeCL: GPU Kernels in Rust for CUDA, ROCm, and WGPU by nathanielsimard on Thursday April 24, 2025 @08:18AM (#43781819)

The need to build CubeCL came from the Burn deep learning framework (https://github.com/tracel-ai/burn), where we want to easily build algorithms like in CUDA with a real programming language, while also being able to integrate those algorithms inside a compiler at runtime to fuse dynamic graphs.

Since we don't want to rewrite everything multiple times, it also has to be multi-platform and optimal, so the feature set must be per-device, not per-language. I'm not aware of a tool that does that, especially in Rust (which Burn is written in).

[ Parent | on HN ]

Re:CubeCL: GPU Kernels in Rust for CUDA, ROCm, and WGPU by fc417fc802 on Thursday April 24, 2025 @04:35PM (#43787190)

> I'm not aware of a tool that does that

Jax? But then you're stuck in python. SYCL?

But yeah not for Rust. This project is filling a prominent hole IMO.

[ Parent | on HN ]

Re:CubeCL: GPU Kernels in Rust for CUDA, ROCm, and WGPU by rowanG077 on Friday April 25, 2025 @01:34PM (#43796408)

Futhark immediately came to mind. It's designed to be able to be trivially integrated into a package.

[ Parent | on HN ]