You can now write CUDA kernels in plain Rust via cuda-oxide and cutile-rs
rjurney · x · 2026-09-10
New projects enable writing CUDA kernels in plain Rust, with two approaches: cuda-oxide (SIMT) and cutile-rs (Tile). The open-source release is circulating in the community, offering Rust developers an alternative to the C++ CUDA toolchain.
Related event: NVIDIA Launches CUDA Rust with Two Paths for Writing GPU Kernels(5 posts)→
More from Infra
- Encrypted GPT-2 speedrun challenges FHE inference: best is 3.031 s/token, goal sub-1s — tensorqt · 2026-10-27
- Open-source Latch compiles PyTorch into training megakernels, 2.75x faster on H100 — sguada · 2026-10-27
- NVIDIA claims HGX Rubin NVL8 delivers 10x tokens per megawatt vs B200 — thursdai_pod · 2026-10-27
- NVIDIA Claims HGX Rubin NVL8 Delivers 10x More Tokens per Megawatt Than B200 — thursdai_pod · 2026-10-27
- Together lets multiple LoRA post-training experiments share the same GPU capacity — togethercompute · 2026-10-27
- Commoditization dilemma: either human labor or GPUs become the commodity — _AustinCalvert_ · 2026-10-27