Today's RAG & Vector Databases: Fastest-Growing Projects — June 10, 2026
This week, the RAG & Vector Databases category on GitHub continues to see significant growth, with developers actively contributing to a variety of projects that leverage retrieval-augmented generation for diverse applications such as legal judgment retrieval, team collaboration, and education. Among these, ZJunCher/xiaoyan-ai-dev-assistant leads the pack with its robust development activity and high user engagement, reflecting the growing interest in RAG-based solutions.
ZJunCher/xiaoyan-ai-dev-assistant is an AI research assistant that combines retrieval-augmented generation (RAG) and multi-round memory for team knowledge sharing and learning. With a growth score of 12.50 and over 100 stars, this project's rapid development and active community indicate its potential as a versatile tool for both experienced developers and newcomers to RAG.
aa0101181514/tw-legal-rag is an open-source command-line interface (CLI) designed for semantic retrieval of Taiwan legal judgments. Users can search, package, and perform citation checks on these judgments using their own LLMs like ChatGPT or Claude. With 9.92 growth score and 171 stars, this tool's steady development and wide appeal among legal professionals suggest its utility in the field.
biao994/DocPaws is an engineering-oriented RAG document assistant that features knowledge base management, PDF indexing, scope-based retrieval, citation tracing, and more. Built with FastAPI and Vue3, it garners a growth score of 9.60 and over 102 stars, highlighting its comprehensive feature set and increasing user interest.
StarTrail-org/PixelRAG marks the end of web parsing by introducing scalable pixel-native search capabilities. With a growth score of 9.54 and 46 stars, this project's innovative approach to visual data retrieval is drawing attention from developers looking for advanced image search solutions.
superman1006/MeetMind transforms requirements into live debates among five role-playing agents (architect, backend, frontend, test, PM) each equipped with its own RAG knowledge base. It supports various LLMs and garners a growth score of 8.07 along with 41 stars, indicating growing interest in collaborative development environments that leverage AI for better decision-making.
qixinhu11/LongLive-RAG presents an implementation of LongLive-RAG, a general framework designed to enhance long video generation through retrieval-augmented methods. With a growth score of 5.15 and 55 stars, this project is gaining traction among researchers and developers interested in multimedia content creation.
ather-techie/rag-interview-system compiles over 200 interview questions related to RAG and outlines system design scenarios and architecture patterns. It has accumulated 4.86 growth score and 52 stars, reflecting its value for those preparing for technical interviews or learning about RAG systems in depth.
nils0000shiyong/Kuaida-AI-assistant is an Android application that enhances interview performance using personalized AI assistance based on the user's own experiences and projects. Its low growth score of 1.94 and 22 stars indicate limited engagement, possibly due to its niche focus or recent launch.
GasolSun36/PyRAG focuses on executable multi-hop reasoning for RAG applications but has seen less activity with a growth score of just 1.00 and 23 stars. This suggests it might be in early stages or targeting a more specialized audience.
Overall, these projects showcase the diverse applications and growing interest in retrieval-augmented generation technologies across various domains such as legal, engineering, multimedia, and education.
ZJunCher/xiaoyan-ai-dev-assistant is an AI research assistant that combines retrieval-augmented generation (RAG) and multi-round memory for team knowledge sharing and learning. With a growth score of 12.50 and over 100 stars, this project's rapid development and active community indicate its potential as a versatile tool for both experienced developers and newcomers to RAG.
aa0101181514/tw-legal-rag is an open-source command-line interface (CLI) designed for semantic retrieval of Taiwan legal judgments. Users can search, package, and perform citation checks on these judgments using their own LLMs like ChatGPT or Claude. With 9.92 growth score and 171 stars, this tool's steady development and wide appeal among legal professionals suggest its utility in the field.
biao994/DocPaws is an engineering-oriented RAG document assistant that features knowledge base management, PDF indexing, scope-based retrieval, citation tracing, and more. Built with FastAPI and Vue3, it garners a growth score of 9.60 and over 102 stars, highlighting its comprehensive feature set and increasing user interest.
StarTrail-org/PixelRAG marks the end of web parsing by introducing scalable pixel-native search capabilities. With a growth score of 9.54 and 46 stars, this project's innovative approach to visual data retrieval is drawing attention from developers looking for advanced image search solutions.
superman1006/MeetMind transforms requirements into live debates among five role-playing agents (architect, backend, frontend, test, PM) each equipped with its own RAG knowledge base. It supports various LLMs and garners a growth score of 8.07 along with 41 stars, indicating growing interest in collaborative development environments that leverage AI for better decision-making.
qixinhu11/LongLive-RAG presents an implementation of LongLive-RAG, a general framework designed to enhance long video generation through retrieval-augmented methods. With a growth score of 5.15 and 55 stars, this project is gaining traction among researchers and developers interested in multimedia content creation.
ather-techie/rag-interview-system compiles over 200 interview questions related to RAG and outlines system design scenarios and architecture patterns. It has accumulated 4.86 growth score and 52 stars, reflecting its value for those preparing for technical interviews or learning about RAG systems in depth.
nils0000shiyong/Kuaida-AI-assistant is an Android application that enhances interview performance using personalized AI assistance based on the user's own experiences and projects. Its low growth score of 1.94 and 22 stars indicate limited engagement, possibly due to its niche focus or recent launch.
GasolSun36/PyRAG focuses on executable multi-hop reasoning for RAG applications but has seen less activity with a growth score of just 1.00 and 23 stars. This suggests it might be in early stages or targeting a more specialized audience.
Overall, these projects showcase the diverse applications and growing interest in retrieval-augmented generation technologies across various domains such as legal, engineering, multimedia, and education.