topoteretes/cognee
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory with small models for free
About the project
Open-source AI memory platform for agents: turns documents, code, and conversations into a self-hosted knowledge graph agents can search and reuse across sessions. Runs locally on free small models with no API key required.
Useful for
- Build a shared team memory from docs, tickets, and code to trace links between decisions
- Give an agent long-term memory of past decisions and fixes across runs
- Run a local knowledge graph from your own text without an LLM or API key via cognee-cli
README summarized by DeepSeek V4.1 Flash. Details may be inaccurate.
Why it’s trending
- 14 stars so far today, about 45 expected by the end of the day.
- Over the last two days the pace is 2.3× that of the previous week and a half.
- GitHub Trending Python today: #13, +103 stars.
- About 8 forks a day — people are taking the code.
Stars per day
Bars are daily stars, the line is the usual pace. Red marks spike days.
Numbers
- Total stars
- 31,110
- Today
- 14 · ≈ 45 by evening
- Forks
- 3,111
- Issues and pull requests
- 5,030
- Watchers
- 114
- Language
- Python
- License
- Apache-2.0
- Latest release
- v1.6.1 · September 24, 2026
- Created
- August 16, 2023
- Last push
- September 27, 2026
Star trust
Growth looks organic: forks and discussion are in line with active projects, and stars arrive unevenly, the way people give them.
These are heuristics, not a verdict: we judge by the repository’s behavior, not by a list of stargazers.
Spotted in
- September 28, 2026GitHub Trending Python today: #13, +103 stars
Similar by description
-
1.2
NxcoreAI/NxMem
A context and long-term memory engine for LLM agents: turns conversations and documents into structured facts, manages short- and long-term memory, and assembles cited context packs.
-
3.3
mem0ai/mem0
A long-term memory layer for AI agents and apps: stores user preferences, session state and facts so assistants keep context across conversations. Available as a Python/JS library, self-hosted server and cloud service.
-
4.7
tigerless-labs/agent-memory
Local long-term memory runtime for AI agents: Markdown files as the source of truth, ranked retrieval over a SQLite index, and sleep-time consolidation. One shared store for Claude Code and Codex, no API keys.
-
1.7
vshulcz/deja-vu
A local Go binary that indexes session history from 34 coding agents (Claude Code, Codex, Cursor, etc.) and gives them a shared memory layer via MCP. No LLM, no embeddings, works from files already on disk.
-
9.2
vectorize-io/hindsight
Hindsight is a long-term memory system for AI agents that stores facts and observations rather than just conversation history, reporting state-of-the-art results on the LongMemEval benchmark. It runs as a server…
-
0.9
HelixDB/helix-db
A Rust graph-vector database for knowledge graphs and AI agent memory, combining graph, vector, full-text and KV storage in one system without separate stores.