Triton

Creator

Creator

Seonglae Cho

Created

Created

2023 May 16 4:1

Editor

Editor

Seonglae Cho

Edited

Edited

2023 Nov 16 8:25

Refs

Refs

Convert triton GPU to LLVM

currently triton only supports linux; 3-bit quantization is not supported when using triton

https://github.com/kimbochen/md-blogs/tree/main/triton

https://www.youtube.com/watch?v=AtbnRIzpwho

Triton Notion

Triton Language

TritonGPUToLLVM

https://www.youtube.com/watch?v=AtbnRIzpwho

https://www.youtube.com/watch?v=AtbnRIzpwho

https://github.com/kimbochen/md-blogs/tree/main/triton

API

Python

Supported Platforms:

Linux

Supported Hardware:

NVIDIA GPUs (Compute Capability 7.0+)
Under development: AMD GPUs, CPUs

Documentations

www.jokeren.tech

https://www.jokeren.tech/slides/triton_intel.pdf

https://github.com/kimbochen/md-blogs/tree/main/triton

Welcome to Triton's documentation! - Triton documentation

Triton is an language and compiler for parallel programming. It aims to provide a Python-based programming environment for productively writing custom DNN compute kernels capable of running at maximal throughput on modern GPU hardware.

https://triton-lang.org/

Introducing Triton: Open-Source GPU Programming for Neural Networks

We're releasing Triton 1.0, an open-source Python-like programming language which enables researchers with no CUDA experience to write highly efficient GPU code-most of the time on par with what an expert would be able to produce. Triton makes it possible to reach peak hardware performance with relatively little

Introducing Triton: Open-Source GPU Programming for Neural Networks

https://openai.com/blog/triton/?utm_source=tldrnewsletter

Introducing Triton: Open-Source GPU Programming for Neural Networks

Paper

www.eecs.harvard.edu

http://www.eecs.harvard.edu/~htk/publication/2019-mapl-tillet-kung-cox.pdf

Recommendations

////////