Development repository for the Triton language and compiler

The Triton Inference Server provides an optimized cloud and edge inferencing solution.

Efficient Triton Kernels for LLM Training

Triton is a dynamic binary analysis library. Build your own program analysis tools, automate your reverse engineering, perform software verification or just emulate code.

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

Puzzles for learning Triton

it is a repository to store all slides used by Triton Ho's public presentation and course.

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.

Distributed Compiler based on Triton for Parallel Systems

Triton DataCenter: a cloud management platform with first class support for containers.

A service for autodiscovery and configuration of applications running in containers

🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"

FlagGems is an operator library for large language models implemented in the Triton Language.

The Triton TensorRT-LLM Backend

Triton's Dragon's Mouth Yellowstone gRPC service for high-performance Solana streaming

Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.

Playing with the Tigress software protection. Break some of its protections and solve their reverse engineering challenges. Automatic deobfuscation using symbolic execution, taint analysis and LLVM.

PyTriton is a Flask/FastAPI-like interface that simplifies Triton's deployment in Python environments.

This repository contains tutorials and examples for Triton Inference Server

🚀🚀🚀A collection of some awesome public projects about Large Language Model(LLM), Vision Language Model(VLM), Vision Language Action(VLA), AI Generated Content(AIGC), the related Datasets and Applications.

TurboQuant: Near-optimal KV cache quantization for LLM inference (3-bit keys, 2-bit values) with Triton kernels + vLLM integration

Triton Python, C++ and Java client libraries, and GRPC-generated client examples for go, java and scala.

Triton backend that enables pre-process, post-processing and other logic to be implemented in Python.

Puzzles for learning Triton, play it with minimal environment configuration!

A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.

Trainable fast and memory-efficient sparse attention

Linux kernel module to support Turbo mode and RGB Keyboard for Acer Predator notebook series

Triton Model Analyzer is a CLI tool to help with better understanding of the compute and memory requirements of the Triton Inference Server models.

flash attention tutorial written in python, triton, cuda, cutlass

A curated list of resources for learning and exploring Triton, OpenAI's programming language for writing efficient GPU code.