m_flow

m_flow

A bio-inspired cognitive memory engine that performs reasoning and associative retrieval through graph structures.

MCPDataOpen source
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 4.4k
Source
mcp-github

Overview

M-flow is a bio-inspired cognitive memory engine that adopts a novel Graph RAG paradigm. Instead of simple similarity matching, it performs reasoning and associative retrieval via graph structures. M-flow organizes knowledge into a four-layer conical graph, ranging from abstract summaries to atomic facts. During querying, the system conducts broad searches across multiple granularities, then propagates evidence paths through the graph, ultimately scoring each knowledge unit. This mechanism enables M-flow to deliver more accurate and relevant results when handling complex queries. Ideal for AI applications requiring advanced retrieval and association capabilities.

Capabilities

  • Multi-granularity vector search
  • Graph-based evidence propagation
  • Scoring based on strongest evidence chain
  • Knowledge organized in a four-layer conical graph

Use cases

Complex query processingKnowledge graph constructionContextual understanding and retrievalEvent causality analysis

Setup

Requires: Python 3.10-3.13API Key
pip install m_flow

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

How is M-flow different from traditional RAG systems?

M-flow performs reasoning and associative retrieval through graph structures, rather than relying solely on similarity matching.

What programming languages does M-flow support?

M-flow primarily supports Python 3.10–3.13.

Related skills