SimpleMem

Active
GitHub Python MIT

Description

SimpleMem: Efficient Lifelong Memory for LLM Agents — supports text and multimodal memory for long-term information retention and retrieval.

Key Features

  • Semantic lossless compression for efficient lifelong memory storage and retrieval in LLM agents
  • Multimodal support for text, image, audio, and video memory with unified package
  • Auto-detection backend routing — first method call determines text or multimodal mode automatically
  • MCP server integration for seamless connection with Claude Desktop, Cursor, LM Studio, and Cherry Studio
  • EvolveMem self-evolving memory via LLM-driven closed-loop diagnosis achieving +25.7% improvement on LoCoMo
  • Cross-session memory persistence outperforming Claude-Mem by 64% with structured memory management

Use Cases

💡 Build AI agents that retain context across long conversations and multiple sessions
💡 Store and retrieve multimodal memories including images, audio clips, and video segments
💡 Create personalized AI assistants that learn user preferences over time
💡 Implement long-term knowledge management for research and documentation workflows

Quick Start

Install via pip: `pip install simplemem`. Initialize with `from simplemem import SimpleMem; mem = SimpleMem()`. Call `mem.add_dialogue()` for text memory or `mem.add_image()` for multimodal. For MCP integration, connect to the cloud-hosted server at mcp.simplemem.cloud.

Related Projects

Related Articles