CPU-first, eager-execution tensor/autograd runtime and LLM inference engine written in pure Zig

Litex: The Language Where Mathematics Verifies Itself.

Continuous Thought Machines, because thought takes time and reasoning is a process.

Minimal and annotated implementations of key ideas from modern deep learning research.

A minimalistic framework for transparently training language models and storing comprehensive checkpoints for in-depth learning dynamics research.

A full attention mechanism and transformer in pure go.

Induce brain-like topographic structure in your neural networks

The only reasonable scripting engine for Go.

Org Novelist is a system for writing novel-length fiction using Emacs Org mode.

A machine learning compiler for GPUs, CPUs, and ML accelerators

Implementation for MatMul-free LM.

Borgo is a statically typed language that compiles to Go.

Source code for the Online-Go.com web interface

LLM training in simple, raw C/CUDA

Boardzilla core library

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

k/simple is a bare minimum k interpreter for learning purposes by arthur whitney

TinyGPT-V: Efficient Multimodal Large Language Model via Small Backbones

Simple, minimal implementation of the Mamba SSM in one file of PyTorch.

Google calendar to org-mode file written in Golang.

Domestic Illumination System Control Operator

An efficient C++20 GPU numerical computing library with Python-like syntax

List of papers studying machine learning through the lens of category theory

An interpreter for RASP as described in the ICML 2021 paper "Thinking Like Transformers"

AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.

Generate Go Plan9 assembly from C code

A two-dimensional, hexagonal programming language.

Foundation Models for Genomics & Transcriptomics

Locally run an Assistant-Tuned Chat-Style LLM