# tatsu-lab/stanford_alpaca

Code and documentation to train Stanford's Alpaca models, and generate the data.

Repository: https://github.com/tatsu-lab/stanford_alpaca
Canonical: https://ross.abutalabs.com/products/stanford_alpaca
Homepage: https://crfm.stanford.edu/2023/03/13/alpaca.html
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deep-learning, instruction-following, language-model
Last push: 2024-07-17T09:43:05+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 90
- inputs: {"age_days": 1272, "days_push": 777, "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 30247, forks 3984 (observed 2026-08-28T04:11:54.315741+00:00)

## What it is
Stanford Alpaca is the code and 52K instruction-following dataset used to fine-tune LLaMA 7B into the Alpaca instruction-following model. It includes data generation scripts, a fine-tuning recipe, and weight-diff recovery code, released for research use only.

## Use cases
- fine-tune a llama model to follow instructions
- generate instruction-following training data with self-instruct
- reproduce the alpaca 7b training recipe
- recover alpaca model weights from a weight diff
- get an instruction tuning dataset for llm research

## When to choose
- you want to replicate or study the Alpaca instruction-tuning recipe
- you need a 52K instruction-following dataset for research fine-tuning
- you want to fine-tune LLaMA 7B cheaply on instruction data

## When to avoid
- you need a production or commercial model - the dataset and weights are CC BY NC 4.0, research only
- you want a ready-to-use chatbot - the demo is suspended and safety tuning is absent
- you need actively maintained training code - the project is a research artifact

## Facets
- artifact type: dataset
- maturity: maintenance
- function: llm-training, data-generation, machine-learning
- domain: large-language-models, machine-learning, deep-learning, artificial-intelligence
- platform: python
- tags: instruction-tuning, llama, fine-tuning, self-instruct, research, alpaca, gpu, linux

## Member repositories
- tatsu-lab/stanford_alpaca (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:54.315741+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 2026-08-29T16:52:47.622123+00:00, confidence not recorded.
  - readme: https://github.com/tatsu-lab/stanford_alpaca (fetched 2026-08-28T04:11:54.315741+00:00, sha 2c7f21c7dcff)
  - homepage: https://crfm.stanford.edu/2023/03/13/alpaca.html (fetched 2026-08-29T07:49:43.781970+00:00, sha 4e0554c4a7b9)
- Data as of 2026-08-30T08:39:29.467469+00:00.