NVIDIA/numba-cuda v0.29.0

v0.29.0

Repository: NVIDIA/numba-cuda

Tag: v0.29.0

Published: 2026-03-17T21:46:43Z

Prerelease: no

Release notes:

What's Changed

• Extend dbg.value coverage to loadvar for scalar kernel parameters by @jiel-nv in https://github.com/NVIDIA/numba-cuda/pull/813
• Fix FP8 uint64 cast flake on Windows by @cpcloud in https://github.com/NVIDIA/numba-cuda/pull/829
• Use dbg.declare for scalar kernel parameters by @cpcloud in https://github.com/NVIDIA/numba-cuda/pull/828
• Fix mixed-IR liveness for inline overload DCE by @cpcloud in https://github.com/NVIDIA/numba-cuda/pull/795
• Use cuda-python for nvvm bindings by @brandon-b-miller in https://github.com/NVIDIA/numba-cuda/pull/818
• fix(ci): cudaRoundMode typing failure in FP8 test by @kaeun97 in https://github.com/NVIDIA/numba-cuda/pull/834
• Support cuda_bindings FastEnum by @mdboom in https://github.com/NVIDIA/numba-cuda/pull/837
• Support cuda.core.GraphBuilder as a kernel-launch stream by @Andy-Jost in https://github.com/NVIDIA/numba-cuda/pull/836
• fix: normalize numpy integer types to python int to prevent overflow errors by @kaeun97 in https://github.com/NVIDIA/numba-cuda/pull/774
• Bump Version to 0.29.0 by @isVoid in https://github.com/NVIDIA/numba-cuda/pull/838

New Contributors

• @Andy-Jost made their first contribution in https://github.com/NVIDIA/numba-cuda/pull/836

Full Changelog: https://github.com/NVIDIA/numba-cuda/compare/v0.28.2...v0.29.0