nanoeuler

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

neural-network

View on GitHub

110 Stars

19 Forks

2 Watchers

Cuda Language

mit License

100 SrcLog Score

Cost to Build

$536.1K

Market Value

$1.53M

How is this calculated?

Growth over time

1 data points · 2026-08-09 → 2026-08-09

Stars Forks Watchers

💬

How do you feel about this project?

Ask AI about nanoeuler

Question copied to clipboard

What is the JustVugg/nanoeuler GitHub project? Description: "GPT-2-style LLM built from scratch in C/CUDA with hand-written backprop, BPE tokenizer, FlashAttention, pretraining, and SFT.". Written in Cuda. Explain what it does, its main use cases, key features, and who would benefit from using it.

Question is copied to clipboard — paste it after the AI opens.

How to clone nanoeuler

Clone via HTTPS

git clone https://github.com/JustVugg/nanoeuler.git

Clone via SSH

[email protected]:JustVugg/nanoeuler.git

Download ZIP

Download main.zip

Found an issue?

Report bugs or request features on the nanoeuler issue tracker:

Open GitHub Issues

Similar to nanoeuler

tensorflow pytorch CNTK spaCy awesome-deep-learning python-machine-learning-book handson-ml tfjs tflearn HanLP tfjs-core Machine-Learning-Tutorials EffectiveTensorflow gensim php-ml brain.js TensorFlow-Tutorials have-fun-with-machine-learning tiny-dnn neurojs TensorLayer Paddle-Lite chainer seq2seq awesome-artificial-intelligence onnx tutorials TensorFlow-World neon tensorflow_cookbook