MyLLM

"LLM from Zero to Hero: An End-to-End Large Language Model Journey from Data to Application!"

Stars: 128

Visit

MyLLM is a web application designed to help law students and legal professionals manage their LLM (Master of Laws) program. It provides features such as course tracking, assignment management, exam schedules, and collaboration tools. Users can easily organize their academic tasks, stay on top of deadlines, and communicate with peers and professors. MyLLM aims to streamline the LLM experience and enhance productivity for students pursuing advanced legal studies.

README:

🚀 MyLLM: Building My Meta_Bot — From Scratch, For Real

⚠️ Work In Progress — Hack at Your Own Risk 🚧

MyLLM isn’t just another library; it's a playground for learning and building LLMs from scratch. This project was born out of a desire to fully understand every line of a transformer stack, from tokenization to RLHF.

Here's what's inside right now:

Area	Status	Description
Interactive Notebooks	✅ Stable	Step-by-step guided learning path
Modular Mini-Projects	✅ Stable	Self-contained, targeted experiments
MyLLM Core Framework	⚙️ Active Development	Pure PyTorch, lightweight, transparent
MetaBot	🛠 Coming Soon	A chatbot that explains itself

Warning: Some parts are stable, while others are actively evolving.

Use this repo to explore, experiment, and break things safely — that's how you learn deeply.

🌱 Why MyLLM Exists

There are plenty of libraries out there (Hugging Face, Lightning, etc.), but they hide too much of the magic. I wanted something different:

Minimal – No unnecessary abstractions, no magic.
Hackable – Every part of the stack is visible and editable.
Research-Friendly – A place to experiment with cutting-edge techniques like LoRA, QLoRA, PPO, and DPO.
From Scratch – So you truly understand the internals.

This is a framework for engineers who want to think like researchers and researchers who want to ship real systems.

🗺 The Three Layers of MyLLM

MyLLM is structured into three progressive layers, designed to guide you from fundamental understanding to building a complete system.

1️⃣ Interactive Notebooks — Learn by Doing

The notebooks/ directory is where your journey begins. Each notebook is a step-by-step guide with theory and code, building components from first principles.

MyLLM/
 └── notebooks/
      ├── 1.DATA.ipynb            # Text preprocessing & tokenization
      ├── 2.ATTENTION.ipynb       # Building the core attention mechanism
      ├── 3.TRAINING.ipynb        # Basic training loop
      ├── 4.FINETUNING.ipynb      # LoRA, QLoRA, and SFT
      ├── 5.RLHF.ipynb            # PPO and DPO algorithms
      └── 6.INFERENCE.ipynb       # KV caching and quantization

💡 Modify the attention mask in a notebook and see how the output changes — that's hands-on learning at its best.

2️⃣ Modular Mini-Projects — Targeted Experiments

The Modules/ folder is a collection of self-contained experiments, each focusing on a specific part of the LLM pipeline. This lets you experiment on one piece of the puzzle without touching the whole framework.

MyLLM/
 └── Modules/
      ├── 1.data/            # Dataset loading and preprocessing utilities
      ├── 2.models/          # Core model architectures (GPT, Llama)
      ├── 3.training/        # Training scripts and utilities
      ├── 4.finetuning/      # Experiments with SFT, DPO, PPO
      └── 5.inference/       # Inference with quantization and KV caching

Example: Train a small GPT from scratch

python Modules/3.training/train.py --config configs/basic.yml

3️⃣ The MyLLM Core Framework — Hugging Face, But From Scratch

The myllm/ folder is where all the components from the notebooks and mini-projects converge into a production-grade framework. This is the final layer, designed for scaling, research, and deployment.

myllm/
 ├── CLI/             # Command-Line Interface
 ├── Configs/         # Centralized configuration objects
 ├── Train/           # Advanced training engine (SFT, DPO, PPO)
 ├── Tokenizers/      # Production-ready tokenizer implementations
 ├── utils/           # Shared utility functions
 ├── api.py           # RESTful API for model serving
 └── model.py         # The core LLM model definition

Example usage:

from myllm.model import LLMModel
from myllm.Train.sft_trainer import SFTTrainer

# Instantiate a model from the core framework
model = LLMModel()

# Fine-tune with a single line of code
trainer = SFTTrainer(model=model, dataset=my_dataset)
trainer.train()

# Every line here maps to real, visible code — no magic.

🔮 Coming Soon: MetaBot

The final vision is MetaBot — an interactive chatbot built entirely with MyLLM.

A chatbot that not only answers your questions but also shows you exactly how it works under the hood.

Built with:

MyLLM core framework
Gradio for UI
Fully open source, located in the Meta_Bot/ directory.

📍 Roadmap

Status	Milestone	Details
✅	Interactive Notebooks	Learn LLM fundamentals hands-on
✅	Modular Mini-Projects	Build reusable, composable components
⚙️	MyLLM Core Framework	Fine-tuning, DPO, PPO, quantization, CLI, API
🛠	MetaBot + Gradio UI	Interactive chatbot & deployment

⚡ Quick Challenges to Try

Run a notebook → tweak hyperparameters → watch how the model changes.
Build a mini GPT that writes haiku poems.
Add a new trainer to the framework (e.g., a TRL variant).
Quantize a model and measure the speedup in inference.
Fork the repo and contribute a new attention mechanism.

🙌 Inspiration

This project wouldn’t exist without the incredible work of others:

Andrej Karpathy — NanoGPT minimalism
Umar Jamil — Practical LLM tutorials
Sebastian Raschka — Deep transformer insights

🏁 The Vision

The end goal: A transparent, educational, and production-ready LLM stack built entirely from scratch, by and for engineers who want to own every line of their AI system.

Let's strip away the black boxes and build the future of LLMs — together.

📜 License

MIT License

For Tasks:

Click tags to check more tools for each tasks

track courses manage assignments schedule exams collaborate with peers communicate with professors

For Jobs:

law student legal assistant paralegal law clerk legal researcher

Alternative AI tools for MyLLM

Similar Open Source Tools

MyLLM

github

: 128

llm-course

The llm-course repository is a collection of resources and materials for a course on Legal and Legislative Drafting. It includes lecture notes, assignments, readings, and other educational materials to help students understand the principles and practices of drafting legal documents. The course covers topics such as statutory interpretation, legal drafting techniques, and the role of legislation in the legal system. Whether you are a law student, legal professional, or someone interested in understanding the intricacies of legal language, this repository provides valuable insights and resources to enhance your knowledge and skills in legal drafting.

github

: 646

Awesome-RL-for-LRMs

This repository contains a collection of awesome resources for reinforcement learning in language models. It includes tutorials, code implementations, research papers, and tools to help researchers and practitioners explore and apply reinforcement learning techniques in natural language processing tasks. Whether you are a beginner or an expert in the field, this repository aims to provide valuable insights and guidance to enhance your understanding and implementation of reinforcement learning in language models.

github

: 1.2k

RAG-To-Know

RAG-To-Know is a versatile tool for knowledge extraction and summarization. It leverages the RAG (Retrieval-Augmented Generation) framework to provide a seamless way to retrieve and summarize information from various sources. With RAG-To-Know, users can easily extract key insights and generate concise summaries from large volumes of text data. The tool is designed to streamline the process of information retrieval and summarization, making it ideal for researchers, students, journalists, and anyone looking to quickly grasp the essence of complex information.

github

: 58

AReaL

AReaL (Ant Reasoning RL) is an open-source reinforcement learning system developed at the RL Lab, Ant Research. It is designed for training Large Reasoning Models (LRMs) in a fully open and inclusive manner. AReaL provides reproducible experiments for 1.5B and 7B LRMs, showcasing its scalability and performance across diverse computational budgets. The system follows an iterative training process to enhance model performance, with a focus on mathematical reasoning tasks. AReaL is equipped to adapt to different computational resource settings, enabling users to easily configure and launch training trials. Future plans include support for advanced models, optimizations for distributed training, and exploring research topics to enhance LRMs' reasoning capabilities.

github

: 2.7k

evaluation-guidebook

The LLM Evaluation guidebook provides comprehensive guidance on evaluating language model performance, including different evaluation methods, designing evaluations, and practical tips. It caters to both beginners and advanced users, offering insights on model inference, tokenization, and troubleshooting. The guide covers automatic benchmarks, human evaluation, LLM-as-a-judge scenarios, troubleshooting practicalities, and general knowledge on LLM basics. It also includes planned articles on automated benchmarks, evaluation importance, task-building considerations, and model comparison challenges. The resource is enriched with recommended links and acknowledgments to contributors and inspirations.

github

: 1.6k

J.A.R.V.I.S.

J.A.R.V.I.S.1.0 is an advanced virtual assistant tool designed to assist users in various tasks. It provides a wide range of functionalities including voice commands, task automation, information retrieval, and communication management. With its intuitive interface and powerful capabilities, J.A.R.V.I.S.1.0 aims to enhance productivity and streamline daily activities for users.

github

: 115

God-Level-AI

A drill of scientific methods, processes, algorithms, and systems to build stories & models. An in-depth learning resource for humans. This repository is designed for individuals aiming to excel in the field of Data and AI, providing video sessions and text content for learning. It caters to those in leadership positions, professionals, and students, emphasizing the need for dedicated effort to achieve excellence in the tech field. The content covers various topics with a focus on practical application.

github

: 3.5k

Eppie-App

Eppie-App is a mobile application designed to help users manage their daily tasks and improve productivity. The app offers features such as task organization, reminders, and goal setting to assist users in staying organized and on track with their responsibilities. With a user-friendly interface and customizable options, Eppie-App aims to simplify task management and enhance efficiency in users' daily lives.

github

: 147

open-webui-tools

Open WebUI Tools Collection is a set of tools for structured planning, arXiv paper search, Hugging Face text-to-image generation, prompt enhancement, and multi-model conversations. It enhances LLM interactions with academic research, image generation, and conversation management. Tools include arXiv Search Tool and Hugging Face Image Generator. Function Pipes like Planner Agent offer autonomous plan generation and execution. Filters like Prompt Enhancer improve prompt quality. Installation and configuration instructions are provided for each tool and pipe.

github

: 348

unity-mcp

MCP for Unity is a tool that acts as a bridge, enabling AI assistants to interact with the Unity Editor via a local MCP Client. Users can instruct their LLM to manage assets, scenes, scripts, and automate tasks within Unity. The tool offers natural language control, powerful tools for asset management, scene manipulation, and automation of workflows. It is extensible and designed to work with various MCP Clients, providing a range of functions for precise text edits, script management, GameObject operations, and more.

github

: 3.2k

Introduction_to_Machine_Learning

This repository contains course materials for the 'Introduction to Machine Learning' course at Sharif University of Technology. It includes slides, Jupyter notebooks, and exercises for the Fall 2024 semester. The content is continuously updated throughout the semester. Previous semester materials are also accessible. Visit www.SharifML.ir for class videos and additional information.

github

: 1.5k

LLMs-playground

LLMs-playground is a repository containing code examples and tutorials for learning and experimenting with Large Language Models (LLMs). It provides a hands-on approach to understanding how LLMs work and how to fine-tune them for specific tasks. The repository covers various LLM architectures, pre-training techniques, and fine-tuning strategies, making it a valuable resource for researchers, students, and practitioners interested in natural language processing and machine learning. By exploring the code and following the tutorials, users can gain practical insights into working with LLMs and apply their knowledge to real-world projects.

github

: 74

gpt-researcher

GPT Researcher is an autonomous agent designed for comprehensive online research on a variety of tasks. It can produce detailed, factual, and unbiased research reports with customization options. The tool addresses issues of speed, determinism, and reliability by leveraging parallelized agent work. The main idea involves running 'planner' and 'execution' agents to generate research questions, seek related information, and create research reports. GPT Researcher optimizes costs and completes tasks in around 3 minutes. Features include generating long research reports, aggregating web sources, an easy-to-use web interface, scraping web sources, and exporting reports to various formats.

github

: 23.6k

PyRIT

PyRIT is an open access automation framework designed to empower security professionals and ML engineers to red team foundation models and their applications. It automates AI Red Teaming tasks to allow operators to focus on more complicated and time-consuming tasks and can also identify security harms such as misuse (e.g., malware generation, jailbreaking), and privacy harms (e.g., identity theft). The goal is to allow researchers to have a baseline of how well their model and entire inference pipeline is doing against different harm categories and to be able to compare that baseline to future iterations of their model. This allows them to have empirical data on how well their model is doing today, and detect any degradation of performance based on future improvements.

github

: 2.9k

PurpleLlama

Purple Llama is an umbrella project that aims to provide tools and evaluations to support responsible development and usage of generative AI models. It encompasses components for cybersecurity and input/output safeguards, with plans to expand in the future. The project emphasizes a collaborative approach, borrowing the concept of purple teaming from cybersecurity, to address potential risks and challenges posed by generative AI. Components within Purple Llama are licensed permissively to foster community collaboration and standardize the development of trust and safety tools for generative AI.

github

: 3.8k

For similar tasks

MyLLM

github

: 128

For similar jobs

llm-course

github

: 646

MyLLM

github

: 128

lawglance

LawGlance is an AI-powered legal assistant that aims to bridge the gap between people and legal access. It is a free, open-source initiative designed to provide quick and accurate legal support tailored to individual needs. The project covers various laws, with plans for international expansion in the future. LawGlance utilizes AI-powered Retriever-Augmented Generation (RAG) to deliver legal guidance accessible to both laypersons and professionals. The tool is developed with support from mentors and experts at Data Science Academy and Curvelogics.

github

: 89

DISC-LawLLM

DISC-LawLLM is a legal domain large model that aims to provide professional, intelligent, and comprehensive **legal services** to users. It is developed and open-sourced by the Data Intelligence and Social Computing Lab (Fudan-DISC) at Fudan University.

github

: 590

ChatLaw

ChatLaw is an open-source legal large language model tailored for Chinese legal scenarios. It aims to combine LLM and knowledge bases to provide solutions for legal scenarios. The models include ChatLaw-13B and ChatLaw-33B, trained on various legal texts to construct dialogue data. The project focuses on improving logical reasoning abilities and plans to train models with parameters exceeding 30B for better performance. The dataset consists of forum posts, news, legal texts, judicial interpretations, legal consultations, exam questions, and court judgments, cleaned and enhanced to create dialogue data. The tool is designed to assist in legal tasks requiring complex logical reasoning, with a focus on accuracy and reliability.

github

: 7.0k

marly

Marly is a tool that allows users to search for and extract context-specific data from various types of documents such as PDFs, Word files, Powerpoints, and websites. It provides the ability to extract data in structured formats like JSON or Markdown, making it easy to integrate into workflows. Marly supports multi-schema and multi-document extraction, offers built-in caching for rapid repeat extractions, and ensures no vendor lock-in by allowing flexibility in choosing model providers.

github

: 135

docutranslate

Docutranslate is a versatile tool for translating documents efficiently. It supports multiple file formats and languages, making it ideal for businesses and individuals needing quick and accurate translations. The tool uses advanced algorithms to ensure high-quality translations while maintaining the original document's formatting. With its user-friendly interface, Docutranslate simplifies the translation process and saves time for users. Whether you need to translate legal documents, technical manuals, or personal letters, Docutranslate is the go-to solution for all your document translation needs.

github

: 72

lawyer-llama

Lawyer LLaMA is a large language model that has been specifically trained on legal data, including Chinese laws, regulations, and case documents. It has been fine-tuned on a large dataset of legal questions and answers, enabling it to understand and respond to legal inquiries in a comprehensive and informative manner. Lawyer LLaMA is designed to assist legal professionals and individuals with a variety of law-related tasks, including: * **Legal research:** Quickly and efficiently search through vast amounts of legal information to find relevant laws, regulations, and case precedents. * **Legal analysis:** Analyze legal issues, identify potential legal risks, and provide insights on how to proceed. * **Document drafting:** Draft legal documents, such as contracts, pleadings, and legal opinions, with accuracy and precision. * **Legal advice:** Provide general legal advice and guidance on a wide range of legal matters, helping users understand their rights and options. Lawyer LLaMA is a powerful tool that can significantly enhance the efficiency and effectiveness of legal research, analysis, and decision-making. It is an invaluable resource for lawyers, paralegals, law students, and anyone else who needs to navigate the complexities of the legal system.

github

: 751