Today's RAG & Vector Databases: Fastest-Growing Projects — November 22, 2026
Today's the RAG & Vector Databases space, there's a notable uptick in projects that focus on integrating retrieval-augmented generation techniques with various enterprise and personal knowledge management solutions. Among these, NVIDIA Auto Ontology's semantic layer stands out as an innovative approach to building personalized knowledge bases for conversational AI models like Claude and ChatGPT.
The project `gogoymh/auto-ontology-mcp` serves as a semantic layer and personal knowledge base server that supports multiple clients including Claude and ChatGPT without requiring any Large Language Model API keys. Its growth is evident from its high growth score of 26.16 and a substantial number of stars at 519, indicating strong interest among developers looking to enhance their conversational AI capabilities.
The `zaxiu88888/issue-rag` repository leverages RAG over GitHub issues to provide a unique feature set that includes an interval report rather than just the best results. This approach is particularly appealing for users seeking detailed insights from issue data, and its growth score of 16.74 alongside 487 stars reflects its growing popularity among developers interested in advanced query mechanisms.
`liangjianshu/multimodal-image-retrieval` focuses on multimodal image retrieval with a comprehensive guide covering intent recognition, prompt engineering, Qwen LoRA, and more. Its description highlights real demonstrations and training evaluations, which might explain its steady growth score of 8.45 and accumulating 128 stars from developers interested in multimodal search techniques.
The `Golden-66/agentic-rag-kb` project aims to build an enterprise-grade intelligent knowledge base using Agentic RAG, combining hybrid retrieval with reranking and agent-based SSE stream processing for real-time question answering. Although its growth score is lower at 3.67, the project has attracted a modest following of 38 stars from those interested in advanced knowledge management solutions.
Lastly, `phonny40rowle/rag-search` offers a self-hosted retrieval-augmented search API that integrates web search, content extraction, embeddings generation, and answer creation capabilities. Despite having only one commit in the last month, its growth score of 1.76 and modest 51 stars suggest it is still gaining traction among those looking for customizable search solutions.
Overall, these projects showcase a diverse range of approaches to leveraging RAG techniques across various use cases, from personal knowledge bases to enterprise-grade intelligent systems and self-hosted search APIs.
The project `gogoymh/auto-ontology-mcp` serves as a semantic layer and personal knowledge base server that supports multiple clients including Claude and ChatGPT without requiring any Large Language Model API keys. Its growth is evident from its high growth score of 26.16 and a substantial number of stars at 519, indicating strong interest among developers looking to enhance their conversational AI capabilities.
The `zaxiu88888/issue-rag` repository leverages RAG over GitHub issues to provide a unique feature set that includes an interval report rather than just the best results. This approach is particularly appealing for users seeking detailed insights from issue data, and its growth score of 16.74 alongside 487 stars reflects its growing popularity among developers interested in advanced query mechanisms.
`liangjianshu/multimodal-image-retrieval` focuses on multimodal image retrieval with a comprehensive guide covering intent recognition, prompt engineering, Qwen LoRA, and more. Its description highlights real demonstrations and training evaluations, which might explain its steady growth score of 8.45 and accumulating 128 stars from developers interested in multimodal search techniques.
The `Golden-66/agentic-rag-kb` project aims to build an enterprise-grade intelligent knowledge base using Agentic RAG, combining hybrid retrieval with reranking and agent-based SSE stream processing for real-time question answering. Although its growth score is lower at 3.67, the project has attracted a modest following of 38 stars from those interested in advanced knowledge management solutions.
Lastly, `phonny40rowle/rag-search` offers a self-hosted retrieval-augmented search API that integrates web search, content extraction, embeddings generation, and answer creation capabilities. Despite having only one commit in the last month, its growth score of 1.76 and modest 51 stars suggest it is still gaining traction among those looking for customizable search solutions.
Overall, these projects showcase a diverse range of approaches to leveraging RAG techniques across various use cases, from personal knowledge bases to enterprise-grade intelligent systems and self-hosted search APIs.