Today's RAG & Vector Databases: Fastest-Growing Projects — April 18, 2026
Today's the RAG & Vector Databases space, we're seeing a surge in innovative tools that leverage Retrieval-Augmented Generation (RAG) to enhance search capabilities and provide more accurate results. Many of these projects are focused on applying RAG to specific use cases, such as document search and threat intelligence analysis.
FlowElement-ai's m_flow takes the top spot with a growth score of 35.78 and 460 stars, as it offers a graph-based approach to finding similar content, while also identifying relevant information through its M-flow component. Its popularity stems from its ability to tackle complex search queries and provide more accurate results.
Joungminsung's OpenDocuments boasts a growth score of 15.89 and 66 stars, as this open-source RAG tool allows users to connect various document sources, such as GitHub, Notion, and Google Drive, and retrieve answers with cited references. Its self-hosted capabilities using Ollama, OpenAI, or Claude have made it an attractive solution for those seeking a customizable search platform.
Yanhua1010's zero-to-ai-fullstack has garnered attention with its growth score of 12.60 and 147 stars, as this Java backend engineer shares their journey of learning AI full-stack in public, incorporating RAG, pgvector, and Next.js into their project. The transparency and willingness to learn have resonated with the community.
Vixhal-baraiya's pageindex-rag has a growth score of 5.22 and 84 stars, offering a vectorless approach to Retrieval-Augmented Generation (RAG) that focuses on reasoning-based retrieval. Its unique methodology has piqued interest among developers seeking alternative approaches to traditional RAG.
Ais1on's CTI-RAG, with a growth score of 4.71 and 54 stars, integrates knowledge graph and causal reasoning capabilities for Cyber Threat Intelligence (CTI) analysis, providing security analysts with an intelligent threat intelligence tool. Its focus on the specific use case has attracted attention from those in the cybersecurity space.
Nashsu's llm_wiki boasts a growth score of 3.92 and an impressive 1,608 stars, as this cross-platform desktop application transforms documents into an organized knowledge base using RAG. By incrementally building and maintaining a persistent wiki, users can access information more efficiently.
Vbj1808's Dokis has a growth score of 2.30 and 34 stars, offering a lightweight RAG provenance middleware that verifies every claim in an LLM response is grounded in a retrieved source without requiring an additional LLM call. Its ability to provide transparency into the reasoning process behind AI-generated content has sparked interest.
McKern3l's RAGdrag rounds out our list with a growth score of 1.76 and 25 stars, providing a security testing toolkit for RAG pipelines that maps to MITRE ATLAS. Its focus on pipeline security has drawn attention from those concerned about the integrity of their AI systems.
FlowElement-ai's m_flow takes the top spot with a growth score of 35.78 and 460 stars, as it offers a graph-based approach to finding similar content, while also identifying relevant information through its M-flow component. Its popularity stems from its ability to tackle complex search queries and provide more accurate results.
Joungminsung's OpenDocuments boasts a growth score of 15.89 and 66 stars, as this open-source RAG tool allows users to connect various document sources, such as GitHub, Notion, and Google Drive, and retrieve answers with cited references. Its self-hosted capabilities using Ollama, OpenAI, or Claude have made it an attractive solution for those seeking a customizable search platform.
Yanhua1010's zero-to-ai-fullstack has garnered attention with its growth score of 12.60 and 147 stars, as this Java backend engineer shares their journey of learning AI full-stack in public, incorporating RAG, pgvector, and Next.js into their project. The transparency and willingness to learn have resonated with the community.
Vixhal-baraiya's pageindex-rag has a growth score of 5.22 and 84 stars, offering a vectorless approach to Retrieval-Augmented Generation (RAG) that focuses on reasoning-based retrieval. Its unique methodology has piqued interest among developers seeking alternative approaches to traditional RAG.
Ais1on's CTI-RAG, with a growth score of 4.71 and 54 stars, integrates knowledge graph and causal reasoning capabilities for Cyber Threat Intelligence (CTI) analysis, providing security analysts with an intelligent threat intelligence tool. Its focus on the specific use case has attracted attention from those in the cybersecurity space.
Nashsu's llm_wiki boasts a growth score of 3.92 and an impressive 1,608 stars, as this cross-platform desktop application transforms documents into an organized knowledge base using RAG. By incrementally building and maintaining a persistent wiki, users can access information more efficiently.
Vbj1808's Dokis has a growth score of 2.30 and 34 stars, offering a lightweight RAG provenance middleware that verifies every claim in an LLM response is grounded in a retrieved source without requiring an additional LLM call. Its ability to provide transparency into the reasoning process behind AI-generated content has sparked interest.
McKern3l's RAGdrag rounds out our list with a growth score of 1.76 and 25 stars, providing a security testing toolkit for RAG pipelines that maps to MITRE ATLAS. Its focus on pipeline security has drawn attention from those concerned about the integrity of their AI systems.