" Introducing three open source projects: Dify, FastGPT and LangChain-Chachat. These projects use a variety of cutting-edge technologies and have the characteristics of modular design, easy scalability, Docker support, etc., which are very suitable for secondary development. Whether it is automating tasks in applications , building a knowledge base, or building a question and answer system, all have a wide range of uses. "
01
—
Diff
Official address: https://dify.ai/
Open source address:
https://github.com/langgenius/dify
Dify = Do It For You
The front-end of this project uses a React development framework Next.js, the back-end is developed using Python's Web framework Flask, and the middleware uses database PostgreSQL, cache Redis, asynchronous queue celery, and vector database Weaviate.
The software architecture is designed to be relatively standardized, easy to expand, and modular. Supports seamless switching of large models at the bottom level.
Support Docker deployment.
02
—
FastGPT
Official address: https://fastgpt.run/
Open source address
https://github.com/labring/FastGPT
Project technology stack: Front-end NextJs + TypeScript + ChakraUI + document database Mongo + database Postgres (Vector plug-in).
Support Docker deployment.
03
—
LangChain-Chatchat
Open source address:
https://github.com/chatchat-space/Langchain-Chatchat
No official address.
The project was originally Langchain-ChatGLM, a local knowledge base question and answer based on language models such as Langchain and ChatGLM.
The article " Project Implementation Practice|Building an exclusive knowledge base based on ChatGLM2-6B + LangChain has been initially completed " is based on this project.
Support Docker deployment.
Reading recommendations:
Andrew Ng: Opportunities for AI
What are the millions of ChatGPT users doing with it?
Embrace the future and learn AI skills! Follow me and receive free AI learning resources.