Fengshenbang-LM
Chinese NLP toolkit
A comprehensive, user-centered ecosystem of pre-trained NLP models for the Chinese language
Fengshenbang-LM(封神榜大模型)是IDEA研究院认知计算与自然语言研究中心主导的大模型开源体系,成为中文AIGC和认知智能的基础设施。
4k stars
58 watching
378 forks
Language: Python
last commit: 6 months ago aigcchinese-nlpdistributed-trainingmultimodalpretrained-modelspytorchtransformers
Related projects:
Repository | Description | Stars |
---|---|---|
| A repository providing a Chinese version of the GPT2 training code, utilizing BERT tokenizer. | 7,488 |
| An open-source chatbot project built on Solid.js and OpenAI's GPT technology, with features like PWA support and customizable prompts. | 3,200 |
| An experimental application showcasing GPT-4's capabilities through automation and AI-driven workflows. | 2,410 |
| An AI-powered desktop application that enables users to query and generate text, images, audio, and video content across various applications | 3,764 |
| A massive corpus of Chinese text data covering various forms and styles | 3,581 |
| A collection of pre-trained GPT2 models and training scripts for multiple languages, including Chinese. | 1,717 |
| A bilingual chat model for human-like conversation | 15,739 |
| A large-scale, bilingual dialogue language model designed to support downstream development and customization for various applications. | 40,832 |
| Teaches Git version control in 30 days through a tutorial and personal experience | 4,004 |
| A large language model based on the Chinese LLaMA architecture, designed to support complex conversations and applications. | 3,641 |
| An implementation of a conversational AI model integrated with WeChat's messaging platform to provide automated customer support and chat services. | 31,919 |
| A curated resource collection of remote work-related information and resources. | 10,204 |
| A repository providing code and models for research into language modeling and multitask learning | 22,644 |
| A proxy tool designed to bypass internet censorship and restrictions by disguising traffic as ordinary network activity. | 33,104 |