<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Triton — FoundationIndex</title>
    <link>https://foundationindex.ustunozgur.com/s/triton</link>
    <description>A Python-like GPU programming language and compiler created by OpenAI that lets ML engineers write fast custom GPU kernels without dropping down to complex C++ CUDA code.</description>
    <language>en</language>
    <atom:link href="https://foundationindex.ustunozgur.com/s/triton/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Triton · Ecosystem · 2023-03-15</title>
      <link>https://foundationindex.ustunozgur.com/s/triton#triton-pytorch-inductor-2023</link>
      <guid>https://foundationindex.ustunozgur.com/s/triton#triton-pytorch-inductor-2023</guid>
      <pubDate>Wed, 15 Mar 2023 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;PyTorch 2.0 adopted Triton as the primary GPU backend for TorchInductor and torch.compile, making Triton-generated fused kernels the default acceleration path across training and inference workloads.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://pytorch.org/get-started/pytorch-2.0/&quot;&gt;PyTorch 2.0 Release&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Triton · Release · 2021-07-28</title>
      <link>https://foundationindex.ustunozgur.com/s/triton#triton-openai-1-0-2021</link>
      <guid>https://foundationindex.ustunozgur.com/s/triton#triton-openai-1-0-2021</guid>
      <pubDate>Wed, 28 Jul 2021 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;OpenAI open-sourced Triton 1.0, giving Python ML engineers a block-level GPU kernel language that matched hand-tuned cuBLAS and CUDA kernels on matrix multiplications and fused attention operations.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://openai.com/index/triton/&quot;&gt;Introducing Triton: Open-Source GPU Programming for Neural Networks&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Triton · Paper · 2019-06-22</title>
      <link>https://foundationindex.ustunozgur.com/s/triton#triton-mapl-paper-2019</link>
      <guid>https://foundationindex.ustunozgur.com/s/triton#triton-mapl-paper-2019</guid>
      <pubDate>Sat, 22 Jun 2019 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Philippe Tillet, H. T. Kung, and David Cox presented Triton: An Intermediate Language and Compiler for Tiled Neural Network Computations at MAPL &amp;apos;19, introducing a tile-based IR that automates shared-memory management and thread scheduling for dense and fused neural kernels.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://dl.acm.org/doi/10.1145/3315508.3329973&quot;&gt;Triton: An Intermediate Language and Compiler for Tiled Neural Network Computations&lt;/a&gt;&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
