This repository serves as a comprehensive, community-driven list of Large Language Models (LLMs) developed in China. It aims to track the development and landscape of Chinese LLMs, providing details on the originating institution, model name, category, and official website. The target audience includes researchers, developers, and anyone interested in the state of AI development within China.

How It Works

The project maintains a curated table of Chinese LLMs, categorized by their developers (companies, universities, research labs) and their primary applications (e.g., general purpose, code, medical, legal). Each entry includes a link to the model's official website or relevant information, and notes on whether it's open-source or has associated applications. The data is updated through community contributions via GitHub Issues.

Quick Start & Requirements

This is a curated list, not a software package. No installation or execution is required. The primary resource is the README itself, which contains the table of LLMs.

Highlighted Details

• Extensive catalog of over 200 Chinese LLMs, with details on developers and categories.
• Includes both commercial and open-source models, noting specific open-source variants (e.g., ChatGLM, Baichuan, Qwen).
• Cross-references with other "Awesome" lists for LLM benchmarks and open foundation models.
• Provides links to official websites and sometimes specific model details or demos.

Maintenance & Community

The repository is maintained by wgwang and relies on community contributions through GitHub Issues for updates and additions. The author also promotes a WeChat public account "走向未来" for AI-related content.

Licensing & Compatibility

The repository itself is likely under a permissive license (common for "Awesome" lists), but the licensing of the individual LLMs listed varies greatly and is not explicitly detailed for each entry. Users must consult the official sources for each model's licensing terms.

Limitations & Caveats

The list is a snapshot and may not be exhaustive or perfectly up-to-date. The "✔" or "✘" for open-source availability is a simplification; actual usage rights and commercial viability depend on each model's specific license. Some entries lack detailed information or links.