CUDA Tile IR

license

Apache-2.0-WITH-LLVM-exception(OSI)

source

public

core-gated

ungated

Apache-2.0-WITH-LLVM-exception license body confirmed. The repository is public and unarchived and builds the whole product, and the README describes no paid tier, enterprise edition or license-gated build beside it, so source is public and the core ungated.

• https://github.com/NVIDIA/cuda-tile/blob/main/LICENSE.txt recorded 2026-09-02

  LICENSE body states the project is under the Apache License v2.0 with LLVM Exceptions and then runs the verbatim Apache-2.0 text plus the LLVM exception clause - real open source, not a closed-binary wrapper repository.

• https://api.github.com/repos/NVIDIA/cuda-tile recorded 2026-09-02

  Repo metadata for cuda-tile's canonical repository - private false, archived false, fork false.

• https://github.com/NVIDIA/cuda-tile/blob/main/README.md recorded 2026-09-02

  README describes CUDA Tile IR as an MLIR-based IR and compiler infrastructure for CUDA kernel optimization, aligned with the CUDA Toolkit 13.1 release, with no paid tier, enterprise edition or license-gated build beside the published source.

1,021 GitHub stars, which lands in the 1K-10K stars band of the stars scale, level 2. The `cuda-tile` package on PyPI (about 3.7M downloads a month) is cuTile Python, published from the separate nvidia/cutile-python repository and pulled in as a dependency of the CUDA Python stack, so it measures a different project's installs rather than this compiler, which the README builds from source with CMake against an LLVM install. No package channel of its own is published, which leaves the stars scale and its cap of 3.

• https://api.github.com/repos/NVIDIA/cuda-tile recorded 2026-09-02

  Repo metadata - stargazers_count = 1,021 - for NVIDIA/cuda-tile.

One band below the apache-tvm anchor: a full compiler infrastructure - IR, bindings, lowering and conformance testing - narrowed to one vendor's targets (NVIDIA GPUs) rather than apache-tvm's multi-vendor retargetability.

• https://github.com/NVIDIA/cuda-tile/blob/main/README.md recorded 2026-09-02

  README describes CUDA Tile IR as an MLIR-based IR and compiler infrastructure for CUDA kernel optimization, aligned with the CUDA Toolkit 13.1 release, with no paid tier, enterprise edition or license-gated build beside the published source.