Download
CCCL, or CUDA Core Compute Libraries, is a unified repository that consolidates several foundational CUDA C++ libraries into a single, cohesive development platform. It brings together Thrust, CUB, and libcudacxx, which collectively provide high-level abstractions, low-level performance primitives, and a CUDA-compatible standard library for GPU programming. The goal of CCCL is to simplify CUDA development by offering reusable building blocks that enable developers to write efficient and scalable parallel code without starting from scratch. Thrust provides a high-level interface for parallel algorithms, while CUB delivers highly optimized primitives for device-level operations, and libcudacxx ensures compatibility with modern C++ standards. By unifying these components, CCCL reduces duplication and improves developer productivity while maintaining performance across different GPU architectures.
Features
- Unified repository combining Thrust, CUB, and libcudacxx
- High-level parallel algorithms and low-level GPU primitives
- CUDA-compatible C++ standard library implementation
- Optimized performance across GPU architectures
- Support for both host and device code execution
- Extensible building blocks for custom CUDA development
Project Samples
