Today's RAG & Vector Databases: Fastest-Growing Projects — April 16, 2026
The RAG & Vector Databases space continues to heat up, with a surge of interest in open-source tools that leverage retrieval-augmented generation (RAG) and vector databases for more efficient and accurate AI-driven applications. Today's top growth scores reflect the demand for innovative solutions that can effectively integrate multiple data sources and provide transparent, explainable results. Notably, several projects are focusing on self-hosted options, suggesting a desire for greater control over data and models.
OpenDocuments by joungminsung takes the top spot with an impressive Growth Score of 17.35 and 64 stars, as it offers a versatile RAG tool that can connect to various platforms like GitHub, Notion, and Google Drive. Its self-hosted capabilities, leveraging Ollama/OpenAI/Claude, have likely contributed to its rapid growth, as developers seek more flexible and customizable solutions.
Yanhua1010's zero-to-ai-fullstack project boasts a Growth Score of 15.44 and an impressive 145 stars, showcasing the popularity of full-stack AI learning resources that incorporate RAG and pgvector. This Java-based project provides a comprehensive framework for building AI-driven applications with Python, FastAPI, and Next.js, making it an attractive resource for developers looking to expand their skillset.
Vixhal-baraiya's pageindex-rag has gained significant attention, with a Growth Score of 5.67 and 82 stars, likely due to its innovative approach to vectorless RAG. By leveraging reasoning-based retrieval methods, this project offers an alternative to traditional vector database approaches, which may appeal to developers seeking more efficient or specialized solutions.
Nashsu's llm_wiki has achieved a remarkable Growth Score of 5.14 and an impressive 1,388 stars, as it provides a user-friendly desktop application for turning documents into organized knowledge bases using RAG. The project's incremental approach to building and maintaining a persistent wiki from sources has likely resonated with users seeking more practical applications of AI-driven tools.
Ais1on's CTI-RAG framework has garnered attention with a Growth Score of 4.80 and 36 stars, as it addresses the specific needs of Cyber Threat Intelligence (CTI) analysis by integrating knowledge graph and causal reasoning capabilities. Although there have been no recent commits, its focused approach to security analysis suggests continued interest in this specialized area.
Vbj1808's Dokis project offers a lightweight RAG provenance middleware that verifies claims in LLM responses without requiring an LLM call, achieving a Growth Score of 2.36 and 34 stars. Its unique approach to ensuring the accuracy and transparency of AI-driven results has likely contributed to its growth, as developers seek more reliable methods for verifying information.
Lastly, McKern3l's RAGdrag project provides a security testing toolkit specifically designed for RAG pipelines, with a Growth Score of 1.87 and 23 stars. Although it has seen slower growth compared to other projects on this list, its focused approach to addressing pipeline security concerns suggests ongoing interest in this critical area.
OpenDocuments by joungminsung takes the top spot with an impressive Growth Score of 17.35 and 64 stars, as it offers a versatile RAG tool that can connect to various platforms like GitHub, Notion, and Google Drive. Its self-hosted capabilities, leveraging Ollama/OpenAI/Claude, have likely contributed to its rapid growth, as developers seek more flexible and customizable solutions.
Yanhua1010's zero-to-ai-fullstack project boasts a Growth Score of 15.44 and an impressive 145 stars, showcasing the popularity of full-stack AI learning resources that incorporate RAG and pgvector. This Java-based project provides a comprehensive framework for building AI-driven applications with Python, FastAPI, and Next.js, making it an attractive resource for developers looking to expand their skillset.
Vixhal-baraiya's pageindex-rag has gained significant attention, with a Growth Score of 5.67 and 82 stars, likely due to its innovative approach to vectorless RAG. By leveraging reasoning-based retrieval methods, this project offers an alternative to traditional vector database approaches, which may appeal to developers seeking more efficient or specialized solutions.
Nashsu's llm_wiki has achieved a remarkable Growth Score of 5.14 and an impressive 1,388 stars, as it provides a user-friendly desktop application for turning documents into organized knowledge bases using RAG. The project's incremental approach to building and maintaining a persistent wiki from sources has likely resonated with users seeking more practical applications of AI-driven tools.
Ais1on's CTI-RAG framework has garnered attention with a Growth Score of 4.80 and 36 stars, as it addresses the specific needs of Cyber Threat Intelligence (CTI) analysis by integrating knowledge graph and causal reasoning capabilities. Although there have been no recent commits, its focused approach to security analysis suggests continued interest in this specialized area.
Vbj1808's Dokis project offers a lightweight RAG provenance middleware that verifies claims in LLM responses without requiring an LLM call, achieving a Growth Score of 2.36 and 34 stars. Its unique approach to ensuring the accuracy and transparency of AI-driven results has likely contributed to its growth, as developers seek more reliable methods for verifying information.
Lastly, McKern3l's RAGdrag project provides a security testing toolkit specifically designed for RAG pipelines, with a Growth Score of 1.87 and 23 stars. Although it has seen slower growth compared to other projects on this list, its focused approach to addressing pipeline security concerns suggests ongoing interest in this critical area.