nanoGPT

nanoGPT is a minimal open-source codebase released by Andrej Karpathy for training GPT-style language models.

1 article
Last mentioned

nanoGPT is an open-source project released on GitHub by Andrej Karpathy. It implements training and fine-tuning of medium-sized GPT models in a short, readable Python and PyTorch codebase, and includes an example that reproduces a GPT-2 model with 124 million parameters from scratch.

Because the code is compact, it is widely used for learning how language model training works and as a starting point for experiments.

This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.

Articles covering this entry

Unlike a full NanoGPT speedrun, where researchers can change nearly everything except the raw data, this track holds the model architecture and data fixed.


© 2026 AIPOST. All rights reserved.

AIPOST is an AI publication covering practical AI, AI security, performance, startups, health, ethics and industry news. No account is needed, and our privacy policy explains how we handle personal information.