2a7ca101d7
Sparse-checkout from NVIDIA/cccl main branch to complete cccl_upstream: Added: - python/cuda_cccl/ (226 files) — Python bindings for device-level algorithms Critical for muh toolchain: cuda.compute.reduce_into, scan, radix_sort, etc. Includes 204 .py files with full test coverage for all 27 algorithms - ci/ (163 files) — Build/test infrastructure build_cub.sh, test_cub.sh, build_and_test_targets.sh, matrix.yaml Directly maps to our [INFRA-CI] and [INFRA-BUILD] items - .agent/skills/ (7 files) — NVIDIA's own agent skills for CCCL cccl-style/SKILL.md, cccl-test/SKILL.md, sass-diff/SKILL.md - docs/ (491 files) — Official CCCL documentation CI references, CMake guides, Python compute docs, libcudacxx PTX docs - test/ (12 files) — Top-level integration tests (cuda_smoke, stdpar) - Root configs: .clang-format, .clang-tidy, CONTRIBUTING.md, pyproject.toml - CLAUDE.md symlink → AGENTS.md (NVIDIA's standard) cccl_upstream now mirrors full NVIDIA/cccl structure: Before: 42M (cub + thrust + libcudacxx + cudax + c + examples + benchmarks) After: 53M (+python +ci +docs +.agent +test +configs) This completes the CCCL base needed for: - [muh-bench] items: ci/util/build_and_test_targets.sh for targeted builds - [CCCL-verify] items: python/cuda_cccl/tests/ as reference implementations - [CCCL-test] items: ci/test_cub.sh, ci/test_thrust.sh - Agent workflow: .agent/skills/ for consistent style and test patterns
CUDA CCCL Python Package
cuda.cccl
provides a Pythonic interface to the
CUDA Core Compute Libraries.
It provides the following modules:
cuda.compute- Device-level parallel algorithms (reduce, scan, sort, etc.) and iteratorscuda.cccl.headers- Programmatic access to CCCL headers
Installation
Install from PyPI:
pip install cuda-cccl[cu13] # For CUDA 13.x (pip-installed cuda-toolkit)
pip install cuda-cccl[cu12] # For CUDA 12.x (pip-installed cuda-toolkit)
If you already have a CUDA toolkit on your system and do not want pip to
install it, use the sysctk variants:
pip install cuda-cccl[sysctk13] # For CUDA 13.x (system CUDA toolkit)
pip install cuda-cccl[sysctk12] # For CUDA 12.x (system CUDA toolkit)
For a minimal install without Numba (useful when supplying pre-compiled operators):
pip install cuda-cccl[minimal-cu13] # pip-installed cuda-toolkit
pip install cuda-cccl[minimal-sysctk13] # system CUDA toolkit
Install from conda-forge:
conda install -c conda-forge cccl-python
Requirements: Python 3.10+, CUDA Toolkit 12.x or 13.x, NVIDIA GPU with Compute Capability 7.5+
Documentation
For complete documentation, examples, and API reference, visit:
- Full Documentation: nvidia.github.io/cccl/unstable/python
- Repository: github.com/NVIDIA/cccl
- Examples: github.com/NVIDIA/cccl/tree/main/python/cuda_cccl/tests/compute/examples