All Topics
All Topics
Technology
Technology
Design
Design
Programming
Programming
Science
Science
News
News
Gaming
Gaming
Entertainment
Entertainment
Business
Business
Finance
Finance
Sports
Sports
Health
Health
Food
Food
Travel
Travel
Art
Art
Music
Music
Books
Books
Education
Education
Politics
Politics
Personal
Personal
No algorithm. No AI slop. No ads. Just RSS. Pro-human. Indie writers. Real journalism. Open web. Chronological. Hand toasted.

Andrej Karpathy Releases nanochat: Full ChatGPT-Style LLM That Can Be Trained for $100

By

bilsbie

7mo ago· 3 min readenNews

Summary

Andrej Karpathy has released nanochat, a complete ChatGPT-style large language model implementation that can be trained for as little as $100. The project is a full-stack LLM implementation in a single, clean codebase of about 8,000 lines, mostly in Python with PyTorch and some Rust for tokenizer training. It includes training, inference, and a web UI, with Karpathy suggesting renting an 8XH100 NVIDIA node for approximately $24/hour, making 4 hours of training sufficient for the model.

Key quotes

· 3 pulled
This repo is a full-stack implementation of an LLM like ChatGPT in a single, clean, minimal, hackable, dependency-lite codebase.
It's around 8,000 lines of code, mostly Python (using PyTorch) plus a little bit of Rust for training the tokenizer.
Andrej suggests renting a 8XH100 NVIDA node for around $24/ hour to train the model. 4 hours (~$100) is enough
Snippet from the RSS feed
Really interesting new project from Andrej Karpathy, described at length in this discussion post. It provides a full ChatGPT-style LLM, including training, inference and a web Ui, that can be …

You might also wanna read