# aiming-lab/SimpleMem

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

Repository: https://github.com/aiming-lab/SimpleMem
Canonical: https://ross.abutalabs.com/products/simplemem
Language: Python
License: MIT
License Family: permissive
Topics: agent, compression, llm, mcp, memory, rag, retrieval, semantic-search, simplemem, knowledge-graph, lifelong-memory, multimodal, python, audio, video, vision
Last push: 2026-07-24T07:40:38+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 94, release rhythm 72, longevity 17
- inputs: {"age_days": 244, "days_push": 40, "days_rel": 104, "gap_med": 36.0, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3720, forks 392 (observed 2026-08-28T04:08:15.436495+00:00)

## What it is
SimpleMem is a Python library and MCP server that provides efficient lifelong memory for LLM agents, storing, compressing, and retrieving long-term memories with semantic lossless compression. It supports text as well as multimodal inputs (image, audio, video) and integrates with any MCP-compatible client such as Claude Desktop, Cursor, and LM Studio.

## Use cases
- give my llm agent long-term memory across sessions
- add persistent memory to a chatbot
- store and retrieve conversation history for an ai assistant
- compress agent memories without losing semantics
- add multimodal memory for images, audio and video to an agent
- run a memory mcp server for claude desktop or cursor
- build a rag pipeline with semantic memory retrieval

## When to choose
- you need persistent, retrievable long-term memory for LLM agents
- you want MCP integration with clients like Claude Desktop or Cursor
- you need multimodal memory support beyond plain text
- you want a lightweight Python package with MIT licensing

## When to avoid
- you only need short-term in-context memory within a single session
- you need a fully managed hosted memory service rather than self-hosted components
- your project is not Python-based and does not support MCP

## Facets
- artifact type: library
- maturity: active
- function: rag, vector-database, search-engine, mcp, agent-framework, llm-inference, compression
- domain: large-language-models, artificial-intelligence, developer-tools
- platform: python, cross-platform, self-hosted
- tags: lifelong-memory, semantic-search, multimodal, knowledge-graph, agent-memory, mcp-server, memory-management, ai-agents, retrieval-augmented-generation, natural-language-processing

## Member repositories
- aiming-lab/SimpleMem (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:15.436495+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:29:20.113396+00:00, confidence not recorded.
  - readme: https://github.com/aiming-lab/SimpleMem (fetched 2026-08-28T04:08:15.436495+00:00, sha e42317ff161d)
  - registry_pypi: https://pypi.org/pypi/simplemem/json (fetched 2026-08-29T09:23:59.918312+00:00, sha 8066074385f8)
- Data as of 2026-08-30T08:39:29.467469+00:00.