Triton

GPU programming is a critical component in AI system courses, which is notoriously difficult to learn and teach, given its unique …

Tensor multiplication with learned weight matrices is the fundamental building block in deep learning models. These matrices can often …

This paper introduces FASTEN, a cutting-edge library developed to address the computational challenges inherent in Heterogeneous Graph …

This paper introduces two extensions to the popular PyTorch machine learning framework, TorchDynamo and TorchInductor, which implement …

Advancements in deep learning are often associated with increasing model sizes. Training and deploying large models require …

A gentle introduction to the Proton profiler and its updates.

Presented our work on multi-level instrumentation-based profiling for Triton kernels.

Presented our research on debugging and profiling of GPU-accelerated AI applications.

Presented the ongoing work on Proton

Presented the Proton and Interpreter tools in the Triton project.

Presented a talk about Triton and called for contributions to improving the language

Presented the FASTEN work for accelerating segmented matrix multiplication

Review Triton’s Interpreter’s progress and future plans

Went through Proton’s design overview

High-level overview of PyTorch 2.0 and Triton integration

Presented a talk about Triton and requested feedback from Intel engineers

Presented a talk about Triton and called for contributions to improving the language

Presented triton programming language and its next step

Presented triton programming language and a deep learning profiler