Today's RAG & Vector Databases: Fastest-Growing Projects — September 04, 2026
Today's the RAG & Vector Databases space, we see a mix of emerging and established projects capturing significant attention from developers and researchers alike. The self-hosted knowledge platform Utopia continues to lead the pack with its innovative approach to managing structured data over time.
deeplethe/utopia
Utopia is a Rust-based self-hosted knowledge management system that leverages a PostgreSQL backend to create a graph database capable of tracking when specific facts were true. With 3,550 stars and a growth score of 90.66, Utopia's popularity suggests it meets an unmet need for robust, context-aware document management solutions.
rostamlabs/rostam
Rostam is an open-source vector database designed to offer sub-microsecond performance with features like hybrid dense-sparse search and quantization support. Despite its relatively modest 30 stars and a growth score of 12.72, Rostam's continuous development (with over 100 commits in the last month) indicates ongoing interest from developers seeking high-performance vector storage solutions.
makralabs/makra
Makra functions as a memory layer designed to serve real-time structured data derived from web queries to AI agents, enhancing their performance with minimal latency. With 28 stars and a growth score of 5.73, Makra's steady development pace (24 commits in the last month) suggests it is gaining traction among developers looking for efficient ways to integrate web data into AI workflows.
Shuo-Liang-0111/RA-Bench
RA-Bench provides a real-event-anchored benchmarking framework aimed at detecting AI-generated videos, particularly relevant in crisis situations. Although the project has seen limited growth (2.89) and only 103 stars, its niche focus on critical applications like video content verification may explain why it garners interest from specific communities within the security and verification domains.
While these projects represent a diverse range of approaches to RAG and vector database solutions, they collectively demonstrate the growing importance of robust data management and retrieval systems in advancing AI capabilities.
deeplethe/utopia
Utopia is a Rust-based self-hosted knowledge management system that leverages a PostgreSQL backend to create a graph database capable of tracking when specific facts were true. With 3,550 stars and a growth score of 90.66, Utopia's popularity suggests it meets an unmet need for robust, context-aware document management solutions.
rostamlabs/rostam
Rostam is an open-source vector database designed to offer sub-microsecond performance with features like hybrid dense-sparse search and quantization support. Despite its relatively modest 30 stars and a growth score of 12.72, Rostam's continuous development (with over 100 commits in the last month) indicates ongoing interest from developers seeking high-performance vector storage solutions.
makralabs/makra
Makra functions as a memory layer designed to serve real-time structured data derived from web queries to AI agents, enhancing their performance with minimal latency. With 28 stars and a growth score of 5.73, Makra's steady development pace (24 commits in the last month) suggests it is gaining traction among developers looking for efficient ways to integrate web data into AI workflows.
Shuo-Liang-0111/RA-Bench
RA-Bench provides a real-event-anchored benchmarking framework aimed at detecting AI-generated videos, particularly relevant in crisis situations. Although the project has seen limited growth (2.89) and only 103 stars, its niche focus on critical applications like video content verification may explain why it garners interest from specific communities within the security and verification domains.
While these projects represent a diverse range of approaches to RAG and vector database solutions, they collectively demonstrate the growing importance of robust data management and retrieval systems in advancing AI capabilities.