# project-baize/baize-chatbot

Let ChatGPT teach your own chatbot in hours with a single GPU!

Repository: https://github.com/project-baize/baize-chatbot
Canonical: https://ross.abutalabs.com/products/baize-chatbot
Homepage: https://arxiv.org/abs/2304.01196
Language: Python
License: GPL-3.0
License Family: copyleft
Last push: 2024-03-17T18:06:03+00:00

## Health v2 (maintenance only)
Score: 21/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 89
- inputs: {"age_days": 1251, "days_push": 899, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3150, forks 281 (observed 2026-08-28T04:07:46.129237+00:00)

## What it is
Baize is an open-source chat model built on LLaMA using LoRA parameter-efficient tuning, trained on 100k self-chat dialogs generated by ChatGPT talking to itself plus Alpaca data. The repository provides data collection scripts, fine-tuning code, LoRA weight merging, and a chat demo, with 7B/13B/30B model releases for research use only.

## Use cases
- train a custom chatbot on a single GPU
- fine-tune llama with lora for multi-turn dialogue
- generate self-chat training data with chatgpt
- run an open-source chat model locally
- merge lora weights into a huggingface model
- build a healthcare domain chatbot

## When to choose
- you want to fine-tune an open chat model cheaply on one GPU
- you need multi-turn dialogue training data generation pipelines
- you want research-grade LLaMA-based chat models with LoRA weights

## When to avoid
- you need a model for commercial use (weights and data are research-only)
- you need actively maintained tooling or the latest model architectures
- you want production-ready serving infrastructure

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-training, machine-learning, chatbot, data-generation
- domain: large-language-models, machine-learning, artificial-intelligence
- platform: python, cli
- tags: lora, llama, self-chat-data, parameter-efficient-tuning, chat-model, research-only, natural-language-processing, gpu

## Member repositories
- project-baize/baize-chatbot (main) score 21

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:46.129237+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 2026-08-30T07:25:54.143847+00:00, confidence not recorded.
  - readme: https://github.com/project-baize/baize-chatbot (fetched 2026-08-28T04:07:46.129237+00:00, sha f69210481e6a)
  - homepage: https://arxiv.org/abs/2304.01196 (fetched 2026-08-29T09:40:29.617638+00:00, sha b2648e5bd094)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T09:40:29.621472+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T09:40:29.625259+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T09:40:29.627115+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T09:40:29.623384+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.