milanm/AutoGrad-Engine

A complete GPT language model (training and inference) in ~600 lines of pure C#, zero dependencies

48
/ 100
Emerging

This project offers a compact, understandable version of the core technology behind large language models like ChatGPT. It takes a dataset of text, like a list of names, and learns the patterns within it. The result is a model that can generate new text, such as original names, that convincingly mimic the style and characteristics of the input data. This is primarily for developers, educators, and students who want to understand the inner workings of a GPT.

332 stars.

Use this if you are a developer or student who wants to deeply understand the mechanics of a GPT model, including its training and inference processes, using a pure C# implementation with zero external dependencies.

Not ideal if you need a production-ready system for generating text, a tool for fine-tuning large pre-trained models, or a solution that prioritizes performance and scales to massive datasets.

AI-development machine-learning-education algorithm-understanding generative-AI C#-development
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 11 / 25
Community 17 / 25

How are scores calculated?

Stars

332

Forks

38

Language

C#

License

MIT

Last pushed

Feb 14, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/milanm/AutoGrad-Engine"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.