NVIDIA/cuda-python

View on GitHub

CUDA Python: Performance meets Productivity

Stars: 3.4k
Forks: 329
Open beginner issues: 0
Indexed issues: 216
Avg merge: 1d 23h
Merged PRs (30d): 116

Dominant language: Cython
License: Apache-2.0
Last GitHub push: Sep 19, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

216 issues indexed so far

[FEA]: Support CU_LAUNCH_ATTRIBUTE_IGNORE in LaunchConfig Open

cuda.core feature

NVIDIA/cuda-python#2625 · 1 assignee · Opened Aug 14, 2026
Add native Windows ARM64 (WoA) CI testing Open

CI/CD P1

NVIDIA/cuda-python#2622 · 1 assignee · Opened Aug 14, 2026
License clarification about the project Open

triage

NVIDIA/cuda-python#2621 · 1 assignee · Opened Aug 14, 2026
CI: enable Windows ARM64 wheel builds for stable CUDA channels Open

triage

NVIDIA/cuda-python#2616 · 1 assignee · Opened Aug 13, 2026
Pool-backed MemoryResource buffers bypass overridden deallocate() methods Open

bug cuda.core

NVIDIA/cuda-python#2615 · 3 comments · 1 assignee · Opened Aug 13, 2026
Upgrade to Sphinx 9.x Open

documentation P1 triage

NVIDIA/cuda-python#2598 · 1 comment · 1 assignee · Opened Aug 11, 2026
CI doesn't test the latest release of cuda_core against cuda_bindings main Open

cuda.core P1

NVIDIA/cuda-python#2449 · 7 comments · 1 assignee · Opened Jul 29, 2026
CI: Make it possible to do a release dry-run from an untagged commit Open

NVIDIA/cuda-python#2447 · 1 assignee · Opened Jul 29, 2026
cuda.core: expose GraphDefinition cloning Open

cuda.core feature P0

NVIDIA/cuda-python#2438 · 1 assignee · Opened Jul 28, 2026
[FEA]: Support inheritance from Buffer Open

cuda.core triage

Difficulty 2/5 1-3 hours Newbie friendliness 62/100

NVIDIA/cuda-python#2435 · 1 comment · Opened Jul 28, 2026
cuda.core: support multidimensional and array-backed graph memcpy nodes Open

cuda.core feature P1

NVIDIA/cuda-python#2420 · 1 assignee · Opened Jul 24, 2026
cuda.core: support clustered and cooperative kernel graph nodes Open

bug cuda.core feature P1

NVIDIA/cuda-python#2421 · 1 assignee · Opened Jul 24, 2026
Migrate from os.path to pathlib.Path Open

cuda.bindings cuda.core cuda.pathfinder enhancement

NVIDIA/cuda-python#2410 · 7 comments · 2 assignees · Opened Jul 23, 2026
Disallow importing of legacy modules outside of the `legacy_tests` subdir Open

test

NVIDIA/cuda-python#2397 · 1 assignee · Opened Jul 21, 2026
[BUG]: VirtualMemoryResource: four pre-existing defects (grow-rollback access loss, dead fast path, finalizer warnings, handle_type docstring) Open

bug cuda.core

NVIDIA/cuda-python#2388 · 1 comment · 1 assignee · Opened Jul 18, 2026
[FEA] cuda.core: add remaining new NVRTC options to ProgramOptions (CUDA 13.1+) Open

cuda.core feature

NVIDIA/cuda-python#2368 · 1 assignee · Opened Jul 14, 2026
[MNT] cuda.core: device-attribute catch-up for CUDA 13.1/13.3 Open

cuda.core enhancement

NVIDIA/cuda-python#2367 · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: green context / device-resource follow-ons (CUDA 13.0-13.3) Open

cuda.core feature

NVIDIA/cuda-python#2366 · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: support stream batch memory operations (incl. CUDA 13.1 atomic reductions) Open

cuda.core feature

NVIDIA/cuda-python#2364 · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: memory-pool location routing and new pool attributes (CUDA 13.0/13.2) Open

cuda.core feature triage

NVIDIA/cuda-python#2362 · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: expose device atomic-operation capability queries (CUDA 13.0) Open

cuda.core feature

NVIDIA/cuda-python#2361 · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: expose graph and graph-node identity/introspection APIs (CUDA 13.1) Open

cuda.core feature

NVIDIA/cuda-python#2360 · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: support stream re-capture into an existing graph (CUDA 13.3) Open

cuda.core feature P1

NVIDIA/cuda-python#2359 · 2 comments · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: support logical endpoints (CUDA 13.3 unicast/multicast memory sharing) Open

cuda.core feature

NVIDIA/cuda-python#2358 · 1 assignee · Opened Jul 14, 2026
[FEA] cuda.core: adopt nvFatbin for multi-arch fat binary assembly Open

cuda.core feature

NVIDIA/cuda-python#2356 · 1 assignee · Opened Jul 14, 2026
CI: Brain-dump the design doc Open

CI/CD documentation P0

NVIDIA/cuda-python#2346 · 1 assignee · Opened Jul 11, 2026
[BUG]: VMM growth frees an unowned address after a failed adjacent reservation Open

NVIDIA/cuda-python#2345 · 1 assignee · Opened Jul 11, 2026
Support passing structs as kernel `launch()` arguments Open

cuda.core enhancement P1

NVIDIA/cuda-python#2343 · 1 assignee · Opened Jul 11, 2026
CI: Fix `cuda-python` dependency test Open

bug CI/CD P0

NVIDIA/cuda-python#2342 · 1 assignee · Opened Jul 11, 2026
CI: Add `cuda.compute` integration tests to nightly CI Open

CI/CD P1 test triage

NVIDIA/cuda-python#2341 · 1 assignee · Opened Jul 11, 2026
CI: Add nvmath-python integration tests to nightly CI Open

CI/CD P1 test triage

NVIDIA/cuda-python#2340 · 3 comments · 1 assignee · Opened Jul 11, 2026
[BUG]: `check_spdx.py` can miss stale copyright years outside staged-file workflows Open

P1

NVIDIA/cuda-python#2273 · 2 comments · 1 reaction · 1 assignee · Opened Jun 28, 2026
[FEA]: Add SM arch detection to CUDA-Pathfinder for LLVM bitcode files Open

cuda.pathfinder feature

NVIDIA/cuda-python#2200 · 9 comments · 1 assignee · Opened Jun 11, 2026
Investigate `gh-pages` impact on full clone size and time Open

P1

NVIDIA/cuda-python#2197 · 13 comments · 1 assignee · Opened Jun 11, 2026
Branch cleanup Open

triage

NVIDIA/cuda-python#2190 · 11 comments · 1 reaction · 1 assignee · Opened Jun 10, 2026
Support Python 3.15 Open

cuda.bindings cuda.core cuda.pathfinder P0 packaging

NVIDIA/cuda-python#2163 · 3 comments · 1 reaction · 1 assignee · Opened Jun 2, 2026
Enable CodeRabbit auto-review on cuda-python (port CCCL's `.coderabbit.yaml`) Open

CI/CD feature P1 triage

Difficulty 4/5 3-5 days Newbie friendliness 45/100

NVIDIA/cuda-python#2137 · Opened May 25, 2026
[ENH]: Make cuda_bindings UnicodeDecodeError more actionable Open

cuda.bindings enhancement P1

NVIDIA/cuda-python#2122 · 1 assignee · Opened May 21, 2026
[Pitch] Nightly `pytest-randomly` flake finder + AI fixer Open

CI/CD feature test triage

Difficulty 5/5 Over a week Newbie friendliness 35/100

NVIDIA/cuda-python#2088 · 2 comments · Opened May 14, 2026
EPIC: Tests that never run in CI Open

CI/CD cuda.bindings cuda.core cuda.pathfinder EPIC experiment P1 test triage

NVIDIA/cuda-python#2077 · 4 comments · 1 assignee · Opened May 13, 2026
[cuda.core] Add support for host_launch (host callback nodes / host function launches) Open

cuda.core feature P1

NVIDIA/cuda-python#2058 · 3 comments · 1 assignee · Opened May 8, 2026
[cuda.core] Add support for Multicast Objects Open

cuda.core feature P1

NVIDIA/cuda-python#2057 · 4 comments · 1 assignee · Opened May 8, 2026
Measure and track `cuda.core` import time Open

cuda.core P1 performance

NVIDIA/cuda-python#2045 · 2 comments · 1 assignee · Opened May 7, 2026
pixi run build-docs doesn't update content Open

bug triage

NVIDIA/cuda-python#2036 · 3 comments · 1 assignee · Opened May 6, 2026
Relative imports from Cython files in cuda_core cause duplicate imports Open

bug cuda.core

NVIDIA/cuda-python#2023 · 5 comments · 1 assignee · Opened May 5, 2026
Inconsistent lifetime management across `from_handle` / `from_*` APIs Open

cuda.core enhancement P1

NVIDIA/cuda-python#1989 · 10 comments · 1 assignee · Opened Apr 29, 2026
[BUG]: `actions/labeler` interference with `restricted-paths-guard.yml` workflow Open

bug CI/CD triage

NVIDIA/cuda-python#1967 · 5 comments · 1 assignee · Opened Apr 23, 2026
[FEA]: Type stubs / .pyi files for cuda.bindings and cuda.core Open

cuda.bindings enhancement P1 packaging

NVIDIA/cuda-python#1963 · 11 comments · 1 assignee · Opened Apr 22, 2026
[FEA]: `if_not_found` flag for all pathfinder APIs? Open

cuda.pathfinder feature

NVIDIA/cuda-python#1954 · 1 comment · 1 assignee · Opened Apr 20, 2026
[FEA]: pathfinder "all_must_work" alert mechanism Open

CI/CD cuda.pathfinder feature P1

NVIDIA/cuda-python#1943 · 1 assignee · Opened Apr 17, 2026

NVIDIA/cuda-python

Get new issues in your inbox