GPU-acceleration of Nocturne via Madrona

ldd as a tree

Solve puzzles. Improve your pytorch.

llama3.np is a pure NumPy implementation for Llama 3 model.

Attempt to recreate the core functionality of v1.29.1 Kubernetes in NodeJS, whle being fully compatible with the kubectl CLI

Automated integer hash function discovery

Kolmogorov Arnold Networks

Borgo is a statically typed language that compiles to Go.

Yoga is an embeddable layout engine targeting web standards.

Grok open release

Deep learning accelerator architectures requiring half the multipliers

Code at the speed of thought – Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.

The canonical spec for ulid

A simple, fast, and secure operating system for the cloud.

Run PyTorch models in the browser using ONNX.js

Mamba SSM architecture

VSCode LLVM Compiler Explorer

Yatool is a cross-platform distribution, building, testing, and debugging toolkit focused on monorepositories

Docking Layout Manager for React

Production-ready C++ Asynchronous Framework with rich functionality

A vite plugin for processing html. It is developed based on lodash template

🥉 useful helpers for react-three-fiber

triangulates a 2D polyline into a stroke

low level utils for polyline miter joins

gets miter normals for a 2D polyline

Hackable and optimized Transformers building blocks, supporting a composable construction.

A hilarious HackerNews parody, all-generated by GPT-4

A minimal, resource efficient unikernel for cloud services