Performance portable parallel programming in Python backed by Kokkos
-
Updated
Oct 9, 2026 - Python
#
performance-portability
Here are 4 public repositories matching this topic...
CG-Kit: Code Generation tool-Kit for algorithmic and performance portability
parallel-computing high-performance-computing code-generation performance-portability algorithmic-portability
-
Updated
Jun 9, 2026 - Python
Measuring what a GPU DSL forecloses, at a fixed algorithm: CUDA vs Triton vs NVIDIA Warp vs Gluon on NFA/DFA simulation. Artifact for HPEC 2026.
automata gpu cuda triton nfa domain-specific-language dfa performance-portability hpec research-artifact
-
Updated
Sep 3, 2026 - Python
Does an attention backend chosen on one GPU stay the right choice on another? A measurement study over six NVIDIA GPUs, forward prefill and KV-cache decode.
benchmark reproducible-research gpu cuda pytorch triton attention performance-portability flash-attention
-
Updated
Sep 27, 2026 - Python
Add this topic to your repo
To associate your repository with the performance-portability topic, visit your repo's landing page and select "manage topics."