-
Notifications
You must be signed in to change notification settings - Fork 329
OpenSep 19, 2026
Due by November 10, 2026
•Last updated 59% complete
List view
0 of 60 selected 0 issues of 60 selected
Add
cuda-ccclas a required dependency tocuda-pythoncuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings moduleP1Medium priority - Should doMedium priority - Should dopackagingAnything related to wheels or Conda packagesAnything related to wheels or Conda packagesRFCPlans and announcementsPlans and announcementsStatus: Open.#691 In NVIDIA/cuda-python;- Status: Open.#2345 In NVIDIA/cuda-python;
[FEA] cuda.core: memory-pool location routing and new pool attributes (CUDA 13.0/13.2)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requesttriageNeeds the team's attentionNeeds the team's attentionStatus: Open.#2362 In NVIDIA/cuda-python;Migrate from os.path to pathlib.Path
cuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings modulecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulecuda.pathfinderEverything related to the cuda.pathfinder moduleEverything related to the cuda.pathfinder moduleenhancementAny code-related improvementsAny code-related improvementsStatus: Open.#2410 In NVIDIA/cuda-python;cuda.core: support multidimensional and array-backed graph memcpy nodes
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#2420 In NVIDIA/cuda-python;cuda.core: support clustered and cooperative kernel graph nodes
bugSomething isn't workingSomething isn't workingcuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#2421 In NVIDIA/cuda-python;EPIC: Typing enhancement
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleenhancementAny code-related improvementsAny code-related improvementsEPICSoul of a releaseSoul of a releaseP1Medium priority - Should doMedium priority - Should doStatus: Open.[FEA] cuda.core: adopt nvFatbin for multi-arch fat binary assembly
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestStatus: Open.#2356 In NVIDIA/cuda-python;[FEA] cuda.core: expose graph and graph-node identity/introspection APIs (CUDA 13.1)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestStatus: Open.#2360 In NVIDIA/cuda-python;[FEA] cuda.core: support logical endpoints (CUDA 13.3 unicast/multicast memory sharing)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestStatus: Open.#2358 In NVIDIA/cuda-python;[FEA] cuda.core: support stream batch memory operations (incl. CUDA 13.1 atomic reductions)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestStatus: Open.#2364 In NVIDIA/cuda-python;[FEA] cuda.core: green context / device-resource follow-ons (CUDA 13.0-13.3)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestStatus: Open.#2366 In NVIDIA/cuda-python;[MNT] cuda.core: device-attribute catch-up for CUDA 13.1/13.3
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleenhancementAny code-related improvementsAny code-related improvementsStatus: Open.#2367 In NVIDIA/cuda-python;[FEA] cuda.core: add remaining new NVRTC options to ProgramOptions (CUDA 13.1+)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestStatus: Open.#2368 In NVIDIA/cuda-python;[FEA] cuda.core: expose device atomic-operation capability queries (CUDA 13.0)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestStatus: Open.#2361 In NVIDIA/cuda-python;[FEA]: Define new protocol(s) defining the size of an object in GPU memory
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#646 In NVIDIA/cuda-python;cuda.core: expose GraphDefinition cloning
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP0High priority - Must do!High priority - Must do!Status: Open.#2438 In NVIDIA/cuda-python;[cuda.core] Add support for Multicast Objects
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#2057 In NVIDIA/cuda-python;build(cuda.core): spike scikit-build-core migration
CI/CDCI/CD infrastructureCI/CD infrastructurecuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings modulecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleexperimentDescribes an investigation or measurementDescribes an investigation or measurementpackagingAnything related to wheels or Conda packagesAnything related to wheels or Conda packagesStatus: Draft (not ready).Pool-backed MemoryResource buffers bypass overridden deallocate() methods
bugSomething isn't workingSomething isn't workingcuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleStatus: Open.EPIC: cuda.core C++ integration with RAPIDs & CCCL
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleEPICSoul of a releaseSoul of a releasefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.RFC: Enable per-thread default stream in free-threading builds
cuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings modulefeatureNew feature or requestNew feature or requestP0High priority - Must do!High priority - Must do!RFCPlans and announcementsPlans and announcementsStatus: Open.#133 In NVIDIA/cuda-python;Support Python 3.15
cuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings modulecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulecuda.pathfinderEverything related to the cuda.pathfinder moduleEverything related to the cuda.pathfinder moduleP0High priority - Must do!High priority - Must do!packagingAnything related to wheels or Conda packagesAnything related to wheels or Conda packagesStatus: Open.#2163 In NVIDIA/cuda-python;Make sure CCCL TMA is compiled and use it (very narrowly)
bugSomething isn't workingSomething isn't workingCI/CDCI/CD infrastructureCI/CD infrastructurecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP1Medium priority - Should doMedium priority - Should doStatus: Draft (not ready).Address duplicate import issue found in pypi-installed cuda-core packages
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP2Low priority - Nice to haveLow priority - Nice to havetestImprovements or additions to testsImprovements or additions to testsStatus: Open (in progress).