BigHugger
GH Repository · aiming-lab

SimpleMem

SimpleMem: Efficient Lifelong Memory for LLM Agents — Text & Multimodal

stars
3,765
30-day movement
+31/day
Related entries
60
Connections
1
dockeragent-appPythonvisionvideoaudioagentmemorysemantic-searchknowledge-graphcompressionragretrievalsimplememlifelong-memorymcpmultimodalllmpython

SimpleMem is a Python project providing lifelong memory for LLM agents, covering both text and multimodal content (audio, video, vision). Its topics indicate it combines retrieval, semantic search, knowledge graphs, and compression, and it ships with an MCP integration and Docker tooling under an MIT license.

Reach for it when your LLM agent needs persistent, retrievable memory across a session lifetime, including multimodal inputs.

Use it to

  • Add long-term memory to an LLM agent
  • Index and retrieve multimodal content
  • Expose memory to agents via MCP
  • Run memory service in Docker
  • Build semantic search over agent history

For Developers building agents that need persistent memory

Role
agent-app
Language
Python
Licence
MIT
Forks
394
Open issues
3
Last push
2026-07-24
Latest release
v0.1.0 · 2026-03-10
topicsllm-agentsmemoryretrievalmultimodalknowledge-graphmcp