DFloat11: Lossless LLM Compression for Efficient GPU Inference

Preferred Generation Benchmark

The Optuna MCP Server is a Model Context Protocol (MCP) server to interact with Optuna APIs.

GitHub Action to debug via SSH over a Cloudflare Tunnel. 👩‍💻

WheelNext Website

taurex3 GPU acceleration using cupy

Multi self-hosted runners on single host!

Distributed LLM inference engine for CPU

Tools for STARK proving using cupy

🏆 A ranked list of awesome machine learning Python libraries. Updated weekly.

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

NVIDIA curated collection of educational resources related to general purpose GPU programming.

Python library to use packages in OptunaHub

Adapted version of llama3.np (NumPy) to a CuPy implementation for the Llama 3 model.

Module, Model, and Tensor Serialization/Deserialization

CuPy-compatible initial value problem library based on Scipy

Github Copilot-like LSP code completion server with local LLM powered by llama.cpp

Audits GitHub repositories to highlight data that cannot be automatically migrated using GitHub's migration tools

MLX: An array framework for Apple silicon

A GitHub CLI extension for migrating GitHub Projects between GitHub accounts and products

GLCM in CUDA

Elevation Mapping on GPU.

GitHub Action for creating a GitHub App Installation Access Token

A macOS app for customizing which browser to start

The blessed GitHub Action, for publishing your 📦 distribution files to PyPI, the tokenless way: https://github.com/marketplace/actions/pypi-publish

This is Go library for building GraphQL client with gqlgen

(Temporary fork, to be remvoed)

CUDA Core Compute Libraries