Cognee

Health v2 · maintenance only

Adoption not part of the score

31294 stars · 3150 forks observed · 2026-10-02

What it is AI-extracted, prompt v1, taxonomy v1, 2026-10-05, confidence 0.97

Cognee is an open‑source AI memory platform that provides persistent, long‑term memory for AI agents. It builds a knowledge graph from documents, code and conversations and enables semantic search and retrieval.

Use cases

• Add persistent memory to AI agents so they can recall past interactions
• Create a company‑wide knowledge base that agents can query with natural language
• Enable retrieval‑augmented generation for LLMs using a combined vector and graph store
• Store and search code, documentation, and chat logs for developer assistants
• Build autonomous agents that follow domain‑specific ontologies and rules

When to choose

• You need a self‑hosted, free memory layer for agents without external API costs
• Your application requires both semantic vector search and graph‑based reasoning
• You want to integrate with multiple LLM providers and run on local CPUs
• You prefer an extensible Python SDK with plugins for popular agent frameworks

When to avoid

• You need a fully managed cloud service with built‑in scaling and SLA guarantees
• Your workload relies on GPU‑accelerated embeddings that are not supported locally
• You require a turnkey solution with no configuration of LLM or embedding providers

Facets

library · maturity active

vector-database search-engine rag llm-inference api-framework artificial-intelligence large-language-models python self-hosted memory-platform knowledge-graph agent-memory open-source local-llm retrieval-augmented-generation agent-framework docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="topoteretes/cognee")