# CVI-SZU/Linly

Chinese-LLaMA 1&2、Chinese-Falcon 基础模型；ChatFlow中文对话模型；中文OpenLLaMA模型；NLP预训练/指令微调数据集

Repository: https://github.com/CVI-SZU/Linly
Canonical: https://ross.abutalabs.com/products/linly
Language: Python
License Family: other
Topics: bert, gpt-3, language-model, llama, nlp, zero-shot-learning, chatbot, chatgpt, chinese, chinese-nlp
Last push: 2024-04-14T05:19:19+00:00

## Health v2 (maintenance only)
Score: 21/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 90
- inputs: {"age_days": 1261, "days_push": 871, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3044, forks 222 (observed 2026-08-28T04:07:39.444373+00:00)

## What it is
Linly is a project releasing Chinese-adapted open large language models, including Chinese-LLaMA 1&2, Chinese-Falcon, Linly-OpenLLaMA base models, and the Linly-ChatFlow dialogue model, along with Chinese pretraining and instruction-tuning datasets. It provides reproducible training, evaluation, and quantized inference pipelines for Chinese LLMs.

## Use cases
- download a Chinese LLaMA or Falcon base model for further fine-tuning
- deploy a Chinese chatbot model locally with quantization
- get Chinese instruction-tuning datasets for LLM training
- continue pretraining an English LLM on Chinese corpora
- run a Chinese conversational model on edge devices with llama.cpp
- evaluate Chinese LLM performance on benchmarks like C-Eval

## When to choose
- you need open Chinese base or chat models with reproducible training code
- you want Chinese pretraining and instruction datasets for your own LLM work
- you need quantized Chinese LLM inference on CUDA or edge devices

## When to avoid
- you need a hosted API service rather than self-hosted model weights
- you require a permissive license for all models (some weights are GPL v3)
- you need actively maintained cutting-edge models, as releases stopped in 2024

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: llm-training, llm-inference, nlp, chatbot, rag
- domain: large-language-models, machine-learning, chatbots
- platform: python
- tags: chinese-nlp, llama, falcon, openllama, instruction-tuning, pretraining-corpus, quantization, chatflow, model-weights, natural-language-processing, gpu, linux, docker

## Member repositories
- CVI-SZU/Linly (main) score 21

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:39.444373+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 2026-08-30T07:28:58.260473+00:00, confidence not recorded.
  - readme: https://github.com/CVI-SZU/Linly (fetched 2026-08-28T04:07:39.444373+00:00, sha e7c182ff20e9)
- Data as of 2026-08-30T08:39:29.467469+00:00.