Torchcomms: A modern PyTorch communications API(pytorch.org)

20 points by paladin314159 1 year ago | 2 comments

Helion: A high-level DSL for performant and portable ML kernels(pytorch.org)

143 points by jarbus 11 months ago | 47 comments

Fault Tolerant Llama training(pytorch.org)

50 points by Mougatine 1 year ago | 8 comments

PyTorch Native Architecture Optimization: Torchao(pytorch.org)

157 points by jonbaer 2 years ago | 48 comments

A bug that taught me more about PyTorch than years of using it(elanapearl.github.io)

426 points by bblcla 1 year ago | 78 comments

Blog post recounting how a training loss plateau turned out to be a PyTorch bug, traced through optimizer internals down to GPU kernels.

Accelerating Generative AI with PyTorch II: GPT, Fast(pytorch.org)

285 points by polyrand 2 years ago | 62 comments

PyTorch Internals: Ezyang's Blog(blog.ezyang.com)

418 points by Anon84 1 year ago | 24 comments

Edward Yang's long-form blog essay on PyTorch internals, based on his talk at the PyTorch NYC meetup on May 14, 2019.

Leaving Meta and PyTorch(soumith.ch)

665 points by saikatsg 11 months ago | 162 comments

Blog post by PyTorch co-creator Soumith Chintala announcing his departure from Meta and the PyTorch project, dated November 6, 2025.

Visual introduction to PyTorch(0byte.io)

372 points by 0bytematt 8 months ago | 25 comments

A visual, beginner-friendly tutorial introducing PyTorch, covering tensors, autograd, and building a simple neural network.

SmolGPT: A minimal PyTorch implementation for training a small LLM from scratch(github.com)

415 points by amrrs 1 year ago | 52 comments

GitHub repository for smolGPT, a minimal PyTorch implementation for training a small GPT-style language model from scratch.

PyTorch-native TPU back end for vLLM(github.com)

4 points by zniturah 6 days ago | 0 comments

GitHub repository for vllm-torchtpu, a PyTorch-native TPU backend for the vLLM LLM inference engine.

A bug that taught me more about PyTorch than years of using it(elanapearl.github.io)

426 points by bblcla 1 year ago | 78 comments

Blog post recounting how a training loss plateau turned out to be a PyTorch bug, traced through optimizer internals down to GPU kernels.