# ymcui/Chinese-LLaMA-Alpaca-3

中文羊驼大模型三期项目 (Chinese Llama-3 LLMs) developed from Meta Llama 3

Repository: https://github.com/ymcui/Chinese-LLaMA-Alpaca-3
Canonical: https://ross.abutalabs.com/products/chinese-llama-alpaca-3
Language: Python
License: Apache-2.0
License Family: permissive
Topics: alpaca, large-language-models, llama, llama-2, llama-3, llm, nlp, llama3
Last push: 2026-04-19T00:59:15+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 78, release rhythm 8, longevity 73
- inputs: {"age_days": 1028, "days_push": 137, "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 1979, forks 170 (observed 2026-08-28T04:06:01.605616+00:00)

## What it is
Chinese-LLaMA-Alpaca-3 is a project releasing Chinese-adapted Llama-3 base and instruct large language models, trained via incremental pretraining on large-scale Chinese data and instruction fine-tuning. It also provides training scripts, instruction datasets, and tutorials for local quantized deployment with tools like transformers, llama.cpp, vLLM, and Ollama.

## Use cases
- run a Chinese-language Llama-3 chat model locally
- fine-tune a Chinese instruction-following LLM
- download Chinese Llama-3 base and instruct model weights
- quantize a Chinese LLM to run on a personal CPU or GPU
- get Chinese instruction tuning datasets like alpaca_zh_51k
- deploy a Chinese LLM with vLLM or Ollama

## When to choose
- you need strong Chinese language understanding from an open Llama-3-based model
- you want to fine-tune or continue pretraining a Chinese LLM with provided scripts and data
- you want local, self-hosted deployment of a Chinese chat model

## When to avoid
- you need a general-purpose English-only LLM
- you need the newest model generation rather than Llama-3-based weights
- you want a hosted API service rather than open weights

## Facets
- artifact type: learning-resource
- maturity: active
- function: llm-training, llm-inference, machine-learning, nlp
- domain: large-language-models, artificial-intelligence
- platform: python, cross-platform
- tags: chinese-llm, llama-3, alpaca, instruction-tuning, open-weights, fine-tuning, quantization, natural-language-processing, gpu

## Member repositories
- ymcui/Chinese-LLaMA-Alpaca-3 (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:01.605616+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 2026-08-30T03:04:43.492414+00:00, confidence not recorded.
  - readme: https://github.com/ymcui/Chinese-LLaMA-Alpaca-3 (fetched 2026-08-28T04:06:01.605616+00:00, sha 716c685eba2c)
- Data as of 2026-08-30T08:39:29.467469+00:00.