Skip to content

Introduce fused GEMM + 1-NN primitive using cuTile - #2249

Open
divyegala wants to merge 49 commits into
NVIDIA:mainfrom
divyegala:cutile-python-to-cpp
Open

Introduce fused GEMM + 1-NN primitive using cuTile#2249
divyegala wants to merge 49 commits into
NVIDIA:mainfrom
divyegala:cutile-python-to-cpp

Conversation

@divyegala

@divyegala divyegala commented Jun 17, 2026

Copy link
Copy Markdown
Contributor

This PR adds infrastructure built on top of existing JIT LTO architecture to generate kernels using cutile-python at build time, and embed them in the C++ library to make them callable from C++.

@copy-pr-bot

copy-pr-bot Bot commented Jun 17, 2026

Copy link
Copy Markdown

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

@divyegala divyegala changed the title cuTile Python to CPP embedding example Introduce fused GEMM + 1-NN primitive using cuTile Jun 24, 2026
Comment thread cpp/tests/CMakeLists.txt Outdated
@divyegala

Copy link
Copy Markdown
Contributor Author

/ok to test 2fb6253

@divyegala

Copy link
Copy Markdown
Contributor Author

/ok to test e28ecd9

@divyegala

Copy link
Copy Markdown
Contributor Author

/ok to test 169309d

Comment thread cpp/src/distance/detail/fused_distance_nn/cutile/fused_1nn_kernel.py Outdated
Comment thread cpp/src/cluster/detail/kmeans_balanced.cuh Outdated
Comment thread cpp/src/distance/detail/fused_distance_nn/cutile/fused_1nn_kernel.py Outdated
@divyegala
divyegala requested a review from a team as a code owner August 11, 2026 23:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

feature request New feature or request non-breaking Introduces a non-breaking change

Projects

Status: No status

Development

Successfully merging this pull request may close these issues.

5 participants