What it does
The skill provides an educational GPT implementation of about 300 lines of code, which the publisher describes as readable and easy to modify. It reproduces GPT-2 (124M) trained on OpenWebText. The implementation is attributed to Andrej Karpathy.
It suits anyone who wants to learn the GPT architecture from scratch, working through a transformer model in a short codebase that can be read and changed.
Training runs in one of two setups: on Shakespeare text using a CPU, or on OpenWebText using multiple GPUs.
Limitations and requirements
Training on OpenWebText is described as requiring multiple GPUs. The Shakespeare option is described as running on a CPU.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- AI Templates (aitmpl) first seen 13 Nov 2026 · last seen 19 Nov 2026 · its page on the source · figures reported by the source: aitmpl downloads 38
model-architecture-nanogpt
Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).
v1.0.0As shown on www.aitmpl.com when we last read it