A high-performance inference engine for AI models

A TTS model capable of generating ultra-realistic dialogue in one pass.

Playing around "Less Slow" coding practices in C++ 20, C, CUDA, PTX, & Assembly, from numerics & SIMD to coroutines, ranges, exception handling, networking and user-space IO

Low-bit LLM inference on CPU/NPU with lookup table

C++ template library for high performance SIMD based sorting algorithms

A very fast hashtable

Fast, Flexible and Portable Structured Generation

Inspirational Mapping

Scalene: a high-performance, high-precision CPU, GPU, and memory profiler for Python with AI-powered optimization proposals

Official inference framework for 1-bit LLMs

Korean SAT leader board

match(it): A lightweight single-header pattern-matching library for C++17 with macro-free APIs.

Simple, portable, and self-contained stacktrace library for C++11 and newer

A multi-voice TTS system trained with an emphasis on quality

Visual Studio Code extension providing language support for StarCraft 2 Galaxy Script

Visual Studio Code extension adds support for SC2Layouts language.

Set of libraries to work with StarCraft II map files: Galaxy Script, Triggers scheme, Game Data XML.

SC2 Replay data dump

GameData, UI and Galaxy files

Command-line application to list and extract files from the CASC (Content Addressable Storage Container) used in Blizzard games.

TernGEMM: General Matrix Multiply Library with Ternary Weights for Fast DNN Inference

A browser based Pokémon fangame heavily inspired by the roguelite genre.

DuckDB is an analytical in-process SQL database management system

Korean Nested Named Entity Corpus

OpenKorPos "base" Corpus.

OpenKorPos main repository.

SDLG is an efficient method to accurately estimate aleatoric semantic uncertainty in LLMs