# mem0ai/mem0

Universal memory layer for AI Agents

Repository: https://github.com/mem0ai/mem0
Canonical: https://ross.abutalabs.com/products/mem0
Homepage: https://mem0.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: ai, chatgpt, llm, python, chatbots, rag, application, long-term-memory, memory, memory-management, state-management, ai-agents, agents, genai
Last push: 2026-08-26T17:28:49+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 83
- inputs: {"age_days": 1170, "days_push": 7, "days_rel": 6, "gap_med": 0, "n_releases_24m": 194}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 64121, forks 7504 (observed 2026-08-28T04:12:19.199933+00:00)

## What it is
Mem0 is a memory layer for AI agents and LLM applications that extracts, stores, and retrieves persistent user context across sessions. It ships as an open-source Python/Node SDK plus a managed platform with features like entity linking, hybrid retrieval, and temporal reasoning.

## Use cases
- give my chatbot long-term memory across sessions
- remember user preferences in an AI assistant
- add persistent context to LLM agents
- store and retrieve facts from conversations
- personalize AI responses based on past interactions
- memory layer for multi-agent systems
- reduce token usage by recalling only relevant context

## When to choose
- you're building chatbots or agents that need persistent user memory
- you want drop-in memory infrastructure with Python or Node SDKs
- you need hybrid semantic, keyword, and entity-based memory retrieval
- you want a managed memory service with an open-source fallback

## When to avoid
- you only need simple session state with no cross-session persistence
- you require fully self-hosted advanced features like graph memory or Dream consolidation, which are platform-only
- you want a general-purpose vector database rather than an opinionated memory layer
- your stack doesn't support Python or JavaScript SDKs

## Facets
- artifact type: library
- maturity: active
- function: rag, agent-framework, llm-inference, vector-database, state-management, sdk
- domain: artificial-intelligence, large-language-models, chatbots, developer-tools
- platform: python, cross-platform, cloud, self-hosted
- tags: ai-memory, long-term-memory, memory-layer, personalization, llm-applications, graph-memory, hosted-service, ai-agents, retrieval-augmented-generation, nodejs

## Member repositories
- mem0ai/mem0 (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:19.199933+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 2026-08-29T16:17:25.784532+00:00, confidence not recorded.
  - readme: https://github.com/mem0ai/mem0 (fetched 2026-08-28T04:12:19.199933+00:00, sha a08b32e32725)
  - homepage: https://mem0.ai (fetched 2026-08-28T17:55:38.457001+00:00, sha 4a830bcc8d93)
  - site_page: https://mem0.ai/pricing (fetched 2026-08-28T17:55:38.465830+00:00, sha 359801587da1)
- Data as of 2026-08-30T08:39:29.467469+00:00.