LycheeMem
Active·★ 236·Apache-2.0·Updated 2026-05-15
★ Trending★ Memory & Context
Compact, efficient, and extensible long-term memory for LLM agents.
LycheeMem is a compact memory framework for LLM agents that organizes memory into working, semantic, and procedural stores. It features a four-stage pipeline (WMManager, SearchCoordinator, SynthesizerAgent, ReasoningAgent) with a background ConsolidatorAgent, and supports multi-channel retrieval (FTS, vector, tag, temporal). It integrates via API, OpenClaw plugin, and MCP protocol.
#agent-memory#langgraph#llm-memory#lychee#lycheemem#mcp#mcp-server#mcp-servers
01
Features
01Efficient conversational memory with dual-threshold token budget
02Structured semantic memory with 7 typed MemoryRecords and Record Fusion
03Action-aware search planning and multi-dimensional relevance scoring
04Procedural memory with skill store and HyDE retrieval
05Modular pipeline with synchronous stages and asynchronous background consolidation
02
Compatibility
Python
Python 3.11+
Verified via docs
OpenClaw
OpenClaw Plugin
Verified via docs
MCP
MCP Protocol
Verified via docs
03
Quick start
1
$ git clone https://github.com/LycheeMem/LycheeMem.git
2
$ cd LycheeMem
3
$ pip install -e ".[dev]"
04
Use cases
↳Persistent long-term memory for LLM-based agents
↳Context-aware chatbots with adaptive memory retrieval
↳Automated memory consolidation and skill extraction from conversations
05
Alternatives
letta★ 23.0k
Letta is the platform for building stateful agents: open AI with advanced memory that can learn and self-improve over time.
DesktopCommanderMCP★ 6.1k
This is MCP server for Claude that gives it terminal control, file system search and diff file editing capabilities
FunASR★ 16.6k
Industrial-grade speech recognition toolkit: 170x realtime, 50+ languages, speaker diarization, emotion detection, streaming, and OpenAI-compatible API.
Related searches
Comments
Log in to leave a comment
- HHayden LewisMay 24, 2026
Extensible design means you can add custom memory backends without framework changes.
- HHarper NguyenMay 16, 2026
Good for agents that need memory without adding significant overhead.
- DDylan WilsonApr 19, 2026
Compact, efficient long-term memory for LLM agents — small footprint, good performance.
- SSpencer MartinezApr 5, 2026
Works reliably across long-running agent sessions.