Teaching transformers to play chess

A minimal PyTorch implementation of probabilistic diffusion models for 2D datasets.

Implementation of Denoising Diffusion Probabilistic Model in Pytorch

Depth Pro: Sharp Monocular Metric Depth in Less Than a Second.

Entropy Based Sampling and Parallel CoT Decoding

It's a poorly named metamath verifier

Toy Golang interpreter written in PHP

Interpreter for the perfect programming language

A massively parallel, high-level programming language

A massively parallel, optimal functional runtime in Rust

Kolmogorov Arnold Networks

A minimal GPU design in Verilog to learn how GPUs work from the ground up

Track-Anything is a flexible and interactive tool for video object tracking and segmentation, based on Segment Anything, XMem, and E2FGVI.

Deep learning accelerator architectures requiring half the multipliers

PyTorch pre-trained model for real-time interest point detection, description, and sparse tracking (https://arxiv.org/abs/1712.07629)

ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extraction

It's a core. Made on Twitch.

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

minimalist vector ad

High Quality Resources on GPU Programming/Architecture

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

perfect programming language

A single notebook for fine-tuning GPT-3.5 turbo

The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.

A natural language interface for computers

Inference Llama 2 in one file of pure C

If tinygrad wasn't small enough for you...

A small toy lisp programming language built in typescript

a typescript library that implements a nearly identical API to Karpathy's micrograd

A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API