chibivue is a minimal vuejs/core implementations (reactivity, virtual dom, component runtime, compiler). An online book for building your own Vue.js.

lm-scratch-pytorch - The code is designed to be beginner-friendly, with a focus on understanding the fundamentals of PyTorch and implementing LLMs from scratch,step by step.

FalGland Minecraft Spinoff

FusterCluck-200: Apple Silicon–tuned, 200M-parameter text-only LLM pipeline with 9,216 context. Bootstrap→Pretrain→SFT curriculum, optimized for M3 Pro.

This project is my submission for the exam "Project Work in Architecture and Platform for Artificial Intelligence"

This repository contains the project that I done during the exam of Optimization Method of Data Science at Sapienza University of Rome.

This repository guides you through the process of building a GPT-style Large Language Model (LLM) from scratch using PyTorch. The structure and approach are inspired by the book Build a Large Language Model (From Scratch) by Sebastian Raschka.

This repository contains reinforcement learning projects showcasing Q-Learning and Deep Q-Network (DQN) implementations. It features an autonomous taxi agent and a space invaders agent trained to complete their respective tasks.

This repository implements Machine Learning models from scratch (using only NumPy and basic Python). It covers both regression with Neural Networks and classification with Support Vector Machines (SVM), including binary and multiclass strategies.

The Kat Compiler Collection for the Scratch visual programming language.

VyomAI: state-of-the-art NLP LLM Vision MultiModel transformers implementation into Pytorch

A custom programming language built from scratch, includes full tokenizer, parser, interpeter in Python, variables, loops and more

noisOut is a DDPM-based implementation for generating high-quality images, inspired by Dhariwal & Nichol (2021) and Ho et al. (2020). Using a UNet architecture with Sinusoidal Position Embeddings, the model learns to reverse a diffusion process, progressively removing noise from images.

echoGPT is a minimal GPT implementation for character-level language modeling with 25.4M parameters. Built with PyTorch, it includes multi-head self-attention, feed-forward layers, and position embeddings. Trained on text like tiny_shakespeare.txt to predict the next character.

Explanation on what a CNI (Container Network Interface) is and how to write it from scratch.

Collection of core machine learning and neural network algorithms implemented from scratch using only Python and NumPy. Designed for learning, experimentation, and understanding the fundamentals without relying on high-level libraries.

Scratch project player written in C++ and QML

Qt library for running and rendering Scratch projects

C++ library for building Scratch project players

configurable moe-llama model training and inference built on pytorch