samsja

ray

Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

python-fire

Python Fire is a library for automatically generating command line interfaces (CLIs) from absolutely any Python object.

ipfs

Peer-to-peer hypermedia protocol

jan

Jan is an open source alternative to ChatGPT that runs 100% offline on your computer. Multiple engine support (llama.cpp, TensorRT-LLM)

click

Python composable command line interface toolkit

pre-commit

A framework for managing and maintaining multi-language pre-commit hooks.

rye

a Hassle-Free Python Experience

llama-fs

A self-organizing file system with llama 3

iroh

A toolkit for building distributed applications

ustreamer

µStreamer - Lightweight and fast MJPEG-HTTP streamer

TransformerEngine

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilization in both training and inference.

logfire

Uncomplicated Observability for Python and beyond! 🪵🔥

pacstall

An AUR-inspired package manager for Ubuntu

dstack

An open-source container orchestration engine for running AI workloads in any cloud or data center. https://discord.gg/u8SmfwPpMd

distributed-llama

Tensor parallelism is all you need. Run LLMs on weak devices or make powerful devices even more powerful by distributing the workload and dividing the RAM usage.

runhouse

Write local debuggable Python which traverses your powerful remote infra. Deploy as-is. Unobtrusive, unopinionated, PyTorch-like APIs.

kademlia

A DHT in Python using asyncio

june

PiPPy

Pipeline Parallelism for PyTorch

attorch

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

gcsfs

Pythonic file-system interface for Google Cloud Storage

CommonLoopUtils

CLU lets you write beautiful training loops in JAX.

butterfish

A shell with AI superpowers

litdata

Streamline data pipelines for AI. Process datasets across 1000s of machines, and optimize data for blazing fast model training.

pydantic-argparse

Typed Argument Parsing with Pydantic

subotai

Subotai, a Kademlia based Distributed Hash Table in Rust

pipegoose

Large scale 4D parallelism pre-training for 🤗 transformers in Mixture of Experts *(still work in progress)*

stable-diffusion-webui-simple-manga-maker

It is an Extension feature used in the WebUI for Stable Diffusion. You can create simple comics with it.

sbahn

Pure Rust Distributed Hash Table

kinema

Holistic rescheduling system for Kubernetes to optimize cluster utilization