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

Package virtual environments for redistribution

Package conda environments for redistribution

s6 overlay for containers (includes execline, s6-linux-utils & a custom init)

check links in web documents or full websites

A more intuitive version of du in rust

Read-only mirror of https://gitlab.gnome.org/GNOME/baobab

Achieve state of the art inference performance with modern accelerators on Kubernetes

Run Slurm in Kubernetes

A little word cloud generator in Python

LM engine is a library for pretraining/finetuning LLMs

Accelerating MoE with IO and Tile-aware Optimizations

Camera controls for Linux

Fixing weird quirks and poor defaults

🎮 ⌨ An easy to use tool to change the behaviour of your input devices.

torchcomms: a modern PyTorch communications API

PyTorch Single Controller

apt-fast: A shellscript wrapper for apt that speeds up downloading of packages.

Gets your TFLOPs for your GPU quickly

Open Source Continuous Inference Benchmarking Qwen3.5, DeepSeek, GPTOSS - GB200 NVL72 vs MI355X vs B200 vs GB300 NVL72 vs H100 & soon™ TPUv6e/v7/Trainium2/3

ArcticInference: vLLM plugin for high-throughput, low-latency inference

Intelligent remote directories and files sync

Official code of "StreamBP: Memory-Efficient Exact Backpropagation for Long Sequence Training of LLMs".

Homepage for ProLong (Princeton long-context language models) and paper "How to Train Long-Context Language Models (Effectively)"

YaFSDP: Yet another Fully Sharded Data Parallel

Easily run PyTorch on multiple GPUs & machines

Simplified access to the system clipboard in Emacs.

Cray-LM unified training and inference stack.

Material for gpu-mode lectures

Count the MACs / FLOPs of PyTorch models