ragflow

RAGFlow is an advanced open-source Retrieval-Augmented Generation (RAG) engine that integrates RAG with Agent capabilities to create a powerful context layer for large language models (LLMs). It provides a streamlined RAG workflow suitable for enterprises of all sizes. Users can leverage pre-built agent templates and a converged context engine to efficiently transform complex data into high-fidelity, production-ready AI systems. The server allows for intelligent document understanding, template-based chunking, and compatibility with various data sources, making it a versatile tool for developers.

To install RAGFlow, ensure you have the following prerequisites: a CPU with at least 4 cores, 16 GB of RAM, and 50 GB of disk space. Additionally, you need Docker version 24.0.0 or higher and Docker Compose version 2.26.1 or higher. If you plan to use the code executor feature, you will also need to install gVisor. Once the prerequisites are met, you can start the server by running the command: docker run -i infiniflow/ragflow. Make sure to check and set vm.max_map_count to at least 262144 before starting the server.

If you encounter issues with memory allocation, ensure that your system's vm.max_map_count is set correctly. This setting can be checked and modified using the sysctl command. Remember that changes to vm.max_map_count will reset after a reboot unless you update the /etc/sysctl.conf file. RAGFlow also supports various data synchronization methods, making it adaptable to different data sources.