Information hub for our project training the largest possible historical LLMs.

Stellux operating system is my research operating system project inspired by Symbiote's philosophy of providing runtime privilege level switching for userspace threads.

A simple, performant and scalable Jax LLM!

Awesome resources for GPUs

Fastest kernels written from scratch

A survey of modern quantization formats (e.g., MXFP8, NVFP4) and inference optimization tools (e.g., TorchAO, GemLite), illustrated through the example of Llama-3.1 inference.

Codebase for Cuda Learning

A Quirky Assortment of CuTe Kernels

🍄 Full stack voxel engine

Kernels, of the mega variety :)

Fast CUDA matrix multiplication from scratch

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

Machine Learning Engineering Open Book

Python Sorted Container Types: Sorted List, Sorted Dict, and Sorted Set

gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

Course : Introduction to Computer Systems

Y86 Pipeline Simulator Rust Implementation, employed in PKU's ICS 2024 archlab

NanoGPT (124M) in 90 seconds

The AI framework that adds the engineering to prompt engineering (Python/TS/Ruby/Java/C#/Rust/Go compatible)

🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN

Lexbor is development of an open source HTML Renderer library. https://lexbor.com

Visual understanding and comic explanation benchmark for LLMs

Nano vLLM

Explore the relationships between convex regular-faced polyhedra.

Implementation of a Transformer, but completely in Triton

Online compiler for HIP and NVIDIA® CUDA® code to WebGPU

A benchmark to evaluate language models on questions I've previously asked them to solve.