Welcome to Triton’s documentation!
View page source

Welcome to Triton’s documentation!

Triton is a 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.

Getting Started

Follow the installation instructions for your platform of choice.
Take a look at the tutorials to learn how to write your first Triton program.

Triton API

triton
triton.language
triton.testing
Triton semantics
triton.language.extra.cuda

Gluon

Learn about Gluon’s lower-level programming model in the Gluon overview.
Browse the Gluon tutorials, examples, and API reference.

Triton MLIR Dialects and Ops

Triton MLIR Dialects and Ops

Going Further

Check out the following documents to learn more about Triton and how it compares against other DSLs for DNNs:

Chapter 1: Introduction
Chapter 2: Related Work
Chapter 3: Debugging and Floating-Point Sanitizer (FpSan)