# karpathy/makemore

An autoregressive character-level language model for making more things

Repository: https://github.com/karpathy/makemore
Canonical: https://ross.abutalabs.com/products/makemore
Language: Python
License: MIT
License Family: permissive
Last push: 2024-06-04T19:32:27+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1546, "days_push": 820, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4215, forks 1034 (observed 2026-08-28T04:08:39.163388+00:00)

## What it is
A single-file, hackable PyTorch implementation of an autoregressive character-level language model, ranging from bigrams to a full Transformer. It is primarily an educational tool for learning how language models work by generating text similar to an input file, such as new baby names.

## Use cases
- generate new baby names from a list of names
- learn how transformers and language models work internally
- train a character-level language model on a small text file
- compare bigram, MLP, RNN, LSTM, GRU, and transformer language models
- generate company name ideas from examples

## When to choose
- you want a minimal, readable codebase to study language model fundamentals
- you need lightweight character-level text generation that runs on a laptop
- you are teaching or learning deep learning with paper-backed implementations

## When to avoid
- you need a production-grade, feature-rich training framework
- you want large-scale word-level or multilingual language modeling
- you need pretrained models or extensive tooling and support

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: machine-learning, llm-training, nlp
- domain: deep-learning, large-language-models, education
- platform: python, cross-platform
- tags: pytorch, educational, language-model, transformer, character-level, text-generation, karpathy, natural-language-processing, gpu

## Member repositories
- karpathy/makemore (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:39.163388+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:22:27.080030+00:00, confidence not recorded.
  - readme: https://github.com/karpathy/makemore (fetched 2026-08-28T04:08:39.163388+00:00, sha 2b12f5adcf86)
- Data as of 2026-08-30T08:39:29.467469+00:00.