NVIDIA/libcudacxx

Health v2 · maintenance only

Flags: archived no_license

Adoption not part of the score

2301 stars · 196 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

NVIDIA's libcudacxx is a C++ Standard Library implementation designed to work across both host CPUs and NVIDIA GPUs, providing standard-conforming containers, algorithms, and primitives usable in CUDA device code. The repository has been archived and development has moved to the NVIDIA/cccl (CUDA C++ Core Libraries) repository.

Use cases

• use std::vector and standard algorithms in CUDA device code
• write portable C++ that runs on both CPU and GPU
• synchronize host and device code with standard-style primitives
• migrate CPU C++ code to run on NVIDIA GPUs
• use C++17/20 features in CUDA kernels
• build heterogeneous C++ applications with a standard library API

When to choose

• you need a standard-conforming C++ library usable inside CUDA kernels
• you want to share code between host and device without rewriting
• you are starting a new project and should use its successor in NVIDIA/cccl

When to avoid

• you are starting a new project - use NVIDIA/cccl instead since this repo is archived
• you need GPU support for non-NVIDIA hardware like AMD or Intel
• you need a general-purpose CPU-only C++ standard library

Facets

library · maturity abandoned

concurrency developer-tools gpu-computing developer-tools cross-platform cpp windows cuda gpu-computing standard-library archived cccl heterogeneous-computing gpu linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="NVIDIA/libcudacxx")