younissk/nanoBeard

The simplest, fastest repository for training/finetuning medium-sized GPTs.

Video+code lecture on building nanoGPT from scratch

GPT-2-style LLM built from scratch in C/CUDA with hand-written backprop, BPE tokenizer, FlashAttention, pretraining, and SFT.

LLM training in simple, raw C/CUDA

A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

🧠 Train a 64M-parameter LLM from scratch in just 2h!

Build LLM from scratch

One-click training of your own GPT. Training a GPT has never been easier for beginners. / 一键预训练+SFT一个属于自己的LLM，0基础训练GPT原来可以这么简单？

A 124.6M LLM trained from scratch on 13B tokens on a single RTX 4090 — tokenizer, pretraining, post-training, evaluation, and reproducible inference.