# zjunlp/LightMem

[ICLR 2026] LightMem: Lightweight and Efficient Memory-Augmented Generation

Repository: https://github.com/zjunlp/LightMem
Canonical: https://ross.abutalabs.com/products/lightmem
Language: Python
License: MIT
License Family: permissive
Topics: agent, ai-agents, artificial-intelligence, chatbot, genai, knowledge, large-language-models, lightweight, llm, long-term-memory, memory, memory-management, natural-language-processing, personalization, python, rag, lightmem
Last push: 2026-08-31T00:38:41+00:00

## Health v2 (maintenance only)
Score: 64/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 35, longevity 32
- inputs: {"age_days": 448, "days_push": 3, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1099, forks 106 (observed 2026-09-01T02:14:08.740968+00:00)

## What it is
LightMem is a lightweight Python framework for adding long-term memory to LLM applications and AI agents, providing memory storage, retrieval, and update mechanisms. It works with cloud APIs (OpenAI, DeepSeek) and local model servers (Ollama, vLLM).

## Use cases
- add long-term memory to an llm chatbot
- remember user preferences across conversations
- build a personalized ai agent with memory
- store and retrieve conversation history for llms
- give my agent persistent memory without heavy infrastructure
- manage memory for llm applications

## When to choose
- you want a lightweight, easy-to-integrate memory layer for LLM apps
- you need flexible backend support for cloud and local models
- you want a research-backed memory framework with active development

## When to avoid
- you need a full production memory platform with enterprise features
- you need multimodal or highly structured memory out of the box
- you require a non-Python stack

## Facets
- artifact type: library
- maturity: active
- function: rag, agent-framework, llm-inference, chatbot
- domain: large-language-models, chatbots
- platform: python, cross-platform
- tags: long-term-memory, memory-management, memory-layer, llm-agents, personalization, research-paper, ai-agents, retrieval-augmented-generation, natural-language-processing

## Member repositories
- zjunlp/LightMem (main) score 64

## Provenance
- Observed fields: from GitHub, fetched 2026-09-01T02:14:08.740968+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 2026-08-30T06:46:37.531412+00:00, confidence not recorded.
  - readme: https://github.com/zjunlp/LightMem (fetched 2026-09-01T02:14:08.740968+00:00, sha 1034758cef32)
  - registry_pypi: https://pypi.org/pypi/lightmem/json (fetched 2026-08-29T12:50:03.705130+00:00, sha 9c83c1220021)
- Data as of 2026-08-30T08:39:29.467469+00:00.