LLM Models

Chinese LLMs

• Taiwan LLM
  • GitHub: https://github.com/MiuLab/Taiwan-LLM 
• TAIDE - Trustworthy AI Dialogue Engine
  • GitHub: https://github.com/taide-taiwan 
  • HuggingFace: https://huggingface.co/taide 
• 01.AI - Yi
  • GitHub: https://github.com/01-ai/Yi 
  • HuggingFace: https://huggingface.co/01-ai/ 
• CKIP-Llama-2-7b 是中央研究院詞庫小組(CKIP)開發的開源可商用繁體中文大型語言模型，以商用開源模型Llama-2-7b以及Atom-7b為基礎，再補強繁體中文的處理能力，並對405個可商用的任務檔案同步進行訓練優化，參數量達70億(7 billion)。
  • GitHub: https://github.com/f901107/CKIP-Llama-2-7b 
  • HuggingFace: https://huggingface.co/spaces/ckiplab/CKIP-Llama-2-7b-chat 
• Qwen - 阿里雲通義千問
  • GitHub: https://github.com/QwenLM/Qwen
  • HuggingFace: https://huggingface.co/Qwen  
  • Doc: https://help.aliyun.com/zh/dashscope/create-a-chat-foundation-model?spm=a2c4g.11186623.0.0.20ea4937azFCan 

Code LLMs

• IBM - Granite code model
  • Hugging Face: https://huggingface.co/ibm-granite 
  • GitHub: https://github.com/ibm-granite 
  • IBM開源程式開發專用Granite語言模型，效能超越當前多數開源模型 | iThome
• Codestral - Mistral's first generative AI model for code 
  • HuggingFace: https://huggingface.co/mistralai/Codestral-22B-v0.1
  • Mistral AI推出輕量程式撰寫輔助模型 | iThome 

Benchmark

• PromptBench: A Unified Library for Evaluating and Understanding Large Language Models.