williamberman

ThunderKittens

Tile primitives for speedy kernels

turingas

Assembler for NVIDIA Volta and Turing GPUs

lerobot

🤗 LeRobot: End-to-end Learning for Real-World Robotics in Pytorch

llm.c

LLM training in simple, raw C/CUDA

maxas

Assembler for NVIDIA Maxwell architecture

grok-1

Grok open release

tt-twitch

tenstorrent kernel from twitch

consistencydecoder

Consistency Distilled Diff VAE

milliForth

A FORTH in 340 bytes — the smallest real programming language ever as of yet.

lone

The standalone Linux Lisp

mmap-sync

Rust library for concurrent data access, using memory-mapped files, zero-copy deserialization, and wait-free synchronization.

color_controlnet

A simple extension of Controlnet for color condition

15445-bootcamp

A basic introduction to coding in modern C++.

fast_bert

time_match

ggblas

vectorflow

VectorFlow is a high volume vector embedding pipeline that ingests raw data, transforms it into vectors and writes it to a vector DB of your choice.

minSDXL

Huggingface-compatible SDXL Unet implementation that is readily hackable

InfiniTensor

commavq

commaVQ is a dataset of compressed driving video

linformer

Implementation of Linformer for Pytorch

datacomp

DataComp: In search of the next generation of multimodal datasets

clip-jax

Train vision models using JAX and 🤗 transformers

XNNPACK

High-efficiency floating-point neural network inference operators for mobile, server, and Web

IF

Stable-Diffusion-Latent-Space-Explorer

Codebase for performing various experiments with Stable Diffusion, supported by the diffusers library.

owl-test

won't it be fun if this works? https://github.com/dmarx/the-rest-of-the-fucking-owl

the-algorithm

Source code for Twitter's Recommendation Algorithm

delta

A syntax-highlighting pager for git, diff, grep, and blame output

ml-ane-transformers

Reference implementation of the Transformer architecture optimized for Apple Neural Engine (ANE)