CUTLASS

license

BSD-3-Clause(OSI)

source

public

core-gated

ungated

BSD-3-Clause 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. GitHub reports NOASSERTION because the LICENSE body opens with an NVIDIA copyright line; the body declares SPDX-License-Identifier BSD-3-Clause and is the verbatim three-clause BSD text.

• https://github.com/NVIDIA/cutlass/blob/main/LICENSE.txt recorded 2026-08-18

  LICENSE body carries an NVIDIA copyright line, declares SPDX-License-Identifier BSD-3-Clause and then runs the verbatim three-clause BSD text.

• https://api.github.com/repos/NVIDIA/cutlass recorded 2026-08-18

  Repo metadata - license spdx_id NOASSERTION, private false, archived false, default branch main - for NVIDIA/cutlass.

• https://github.com/NVIDIA/cutlass/blob/main/README.md recorded 2026-08-18

  README describes C++ abstractions for high-performance GEMM and related computations at every level of the CUDA hierarchy, with hierarchical decomposition and data movement specialized through tiling sizes, data types and algorithmic policy, with no paid tier, enterprise edition or license-key-gated build beside it.

10,628 PyPI downloads of `nvidia-cutlass` in the trailing 30 days, which lands in the 10K-100K band of the software usage scale, level 2. CUTLASS is a header-only template library used from source in the great majority of cases, so the wheel figure is a floor rather than a measure. It is still the product's own distribution channel and is banded as recorded.

• https://pypistats.org/api/packages/nvidia-cutlass/recent recorded 2026-08-18

  last_month downloads = 10,628 for nvidia-cutlass

Banded on the category feature matrix as performance-critical kernel and operator library. Placed two bands below the apache-tvm anchor, on a matrix that bands on how much of the model-to-hardware transformation pipeline a product performs, over how many inputs and targets.

• https://github.com/NVIDIA/cutlass/blob/main/README.md recorded 2026-08-18

  README still documents C++ abstractions for high-performance GEMM and related computations at every level of the CUDA hierarchy, with hierarchical decomposition and data movement specialized through tiling sizes, data types and algorithmic policy.