---
id: "nr-nvidia-introduces-cuda-rust-with-two-tracks-for-gpu-f0931a"
slug: "nvidia-introduces-cuda-rust-with-two-tracks-for-gpu-f0931a"
title: "NVIDIA introduces CUDA Rust with two tracks for GPU kernels"
observed_at: "2026-09-18T13:40:45.45798+00:00"
record_date: "2026-09-18"
date_kind: "observed_at"
novelty: "new"
source_platform: "developer.nvidia.com"
topics: ["gpu-programming","developer-tools","ai-infrastructure"]
entities: ["NVIDIA","cuda-oxide","cutile-rs"]
related_patterns: []
source_url: "https://developer.nvidia.com/blog/introducing-cuda-rust-two-tracks-for-writing-gpu-kernels"
source_urls: ["https://developer.nvidia.com/blog/introducing-cuda-rust-two-tracks-for-writing-gpu-kernels"]
basket: {"id":"0325e6b8-f527-4f9b-bfa3-3f30e57f380b","revision":1,"cluster_id":"f0931ac9-648c-439c-b3bb-b12c6827bf65"}
type: "raw_signal"
description: "NVIDIA introduced CUDA Rust, offering cuda-oxide for SIMT kernels and cutile-rs for tile-based kernels, both compiling Rust GPU code without wrapping other languages."
why_it_matters: "For Rust-based inference and serving stacks, kernels can live in the same language as host code, with compile-time checks such as DisjointSlice and launch contracts guarding against aliasing and invalid launches. cuda-oxide is still early alpha, and interoperability with CUDA C++ and CUDA Python is planned rather than shipped."
verification_level: "source-inspected"
signal_type: "newsroom_basket_signal"
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:nvidia-introduces-cuda-rust-with-two-tracks-for-gpu-f0931a"
retrieval_nugget: "NVIDIA introduced CUDA Rust, offering cuda-oxide for SIMT kernels and cutile-rs for tile-based kernels, both compiling Rust GPU code without wrapping other languages. NVIDIA introduced CUDA Rust with two tracks for writing GPU kernels in Rust. cuda-oxide is a custom rustc codegen backend that compiles SIMT-style kernels to PTX and remains in early alpha on a pinned nightly toolchain, while"
status: "published"
visuals: []
routes: {"html":"https://newruntime.com/signals/nvidia-introduces-cuda-rust-with-two-tracks-for-gpu-f0931a/","markdown":"https://newruntime.com/signals/nvidia-introduces-cuda-rust-with-two-tracks-for-gpu-f0931a.md","json":"https://newruntime.com/signals/nvidia-introduces-cuda-rust-with-two-tracks-for-gpu-f0931a.json"}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
---

# NVIDIA introduces CUDA Rust with two tracks for GPU kernels

## Retrieval answer

NVIDIA introduced CUDA Rust, offering cuda-oxide for SIMT kernels and cutile-rs for tile-based kernels, both compiling Rust GPU code without wrapping other languages. NVIDIA introduced CUDA Rust with two tracks for writing GPU kernels in Rust. cuda-oxide is a custom rustc codegen backend that compiles SIMT-style kernels to PTX and remains in early alpha on a pinned nightly toolchain, while

NVIDIA introduced CUDA Rust with two tracks for writing GPU kernels in Rust. cuda-oxide is a custom rustc codegen backend that compiles SIMT-style kernels to PTX and remains in early alpha on a pinned nightly toolchain, while cutile-rs offers tile-based programming on stable Rust 1.89+ with CUDA 13.3 and is published on crates.io.

For Rust-based inference and serving stacks, kernels can live in the same language as host code, with compile-time checks such as DisjointSlice and launch contracts guarding against aliasing and invalid launches. cuda-oxide is still early alpha, and interoperability with CUDA C++ and CUDA Python is planned rather than shipped.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Gpu Programming](https://newruntime.com/topics/gpu-programming.md): Explore the gpu-programming topic hub.
- topic hub: [Developer Tools](https://newruntime.com/topics/developer-tools.md): Explore the developer-tools topic hub.
- topic hub: [Ai Infrastructure](https://newruntime.com/topics/ai-infrastructure.md): Explore the ai-infrastructure topic hub.
- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)