ELS-RD/kernl

Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable.

Jupyter NotebookPythonOthercudapytorchtransformertritoncuda-kernel
This is stars and forks stats for /ELS-RD/kernl repository. As of 02 May, 2024 this repository has 1352 stars and 78 forks.

Kernl lets you run Pytorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable. benchmarks ran on a 3090 RTX Kernl is the first OSS inference engine written in CUDA C OpenAI Triton, a new language designed by OpenAI to make it easier to write GPU kernels. Each kernel is less than 200 lines of code, and is easy to understand and modify. Tutorials - End to End Use Cases A list of Examples contains how to use kernl with Pytorch. Topic Notebook Tiled...
Read on GithubGithub Stats Page
repotechsstarsweeklyforksweekly
sail-sg/metaformerJupyter NotebookPythonShell2910160
NikodemBartnik/Pico-Mars-RovernesCPython920160
Kozea/WeasyPrintPythonCSSHTML6.1k06220
PP-FM/ppsatRoffCPython14050
localstack/docsHTMLSCSSPython38+170-1
stakewithus/defi-by-exampleSolidityJavaScriptPython531+12200
avisalmon/VGAstarter_DE10_liteSystemVerilogPython1020
Simulation-Software-Engineering/Lecture-MaterialPythonTeXShell400430
fduran/sadserversHCLPythonShell9410210
svenlombaert/IBOOK6ActionScriptOther0000