ruflo-ruvector

PackDatabases & data

ruflo-ruvector is a plugin that provides a self-learning vector database. It stores and searches embeddings so an agent can find related information fast using semantic similarity. It combines HNSW, FlashAttention-3, Graph RAG, hybrid search, DiskANN, and Brain AGI in one data layer.

Unavailable. Delivery for this kind is on the roadmap — not serving yet.

Have embeddings for the information you want the agent to search.

What your AI can do with it

  • Stores embeddings for semantic similarity search
  • Indexes vectors with HNSW and DiskANN for fast retrieval
  • Runs hybrid search across vector and other signals
  • Uses FlashAttention-3 for attention-based processing
  • Supports Graph RAG to connect related information
  • Applies Brain AGI for self-learning behavior

Getting started

  1. Have embeddings for the information you want the agent to search.
  2. Install the ruflo-ruvector plugin in your agent setup.
  3. Configure the vector database with your stored embeddings.
  4. Point your agent at the database so it can search by semantic similarity.

Signals

GitHub stars
73k
Forks
9k
Last commit
Sep 2026

Questions

What does ruflo-ruvector do?
It is a plugin providing a self-learning vector database. It stores and searches embeddings so an agent can find related information fast using semantic similarity.
What indexing methods does it use?
It uses HNSW and DiskANN for vector indexing, and supports hybrid search that combines multiple search signals.
What is Brain AGI?
Brain AGI is one of the components listed for ruflo-ruvector, tied to its self-learning behavior. Further details are not specified.
Does it support Graph RAG?
Yes, Graph RAG is listed among its features alongside HNSW, FlashAttention-3, hybrid search, DiskANN, and Brain AGI.
Advanced
Item type
plugin
Key
ruvnet-ruflo-ruflo-ruvector
Source
github.com/ruvnet/ruflo