🌸
- Seoul, Republic of Korea
-
-
flash-attention Public
Forked from ROCm/flash-attentionFast and memory-efficient exact attention
Python BSD 3-Clause "New" or "Revised" License UpdatedJun 17, 2024 -
-
hiprtc-example Public
Use HIP RTC to instantiate templated kernel and invoke it later
C++ UpdatedOct 25, 2022 -
HIP: C++ Heterogeneous-Compute Interface for Portability
C++ MIT License UpdatedOct 2, 2022 -
-
-
-
Tensile Public
Forked from jichangjichang/TensileStretching GPU performance for GEMMs and tensor contractions.
Python MIT License UpdatedMar 8, 2022 -
tokenizer Public
Forked from boostorg/tokenizerBoost.org tokenizer module
C++ Boost Software License 1.0 UpdatedAug 17, 2020