Mem0 review

Q: How do I install Mem0?

Run pip install mem0ai. It needs Node.js >=18. The install section above has the other options from the project's README.

Mem0 is a memory layer for AI agents and apps that stores and retrieves user, session, and agent context across interactions.

Updated Nov 1, 2026. Review written Oct 22, 2026 from the project's own sources; usage data refreshed every Monday.

TL;DR

Mem0 is a memory layer that lets AI assistants and agents remember user preferences and past interactions. Mem0 is for developers building chatbots, assistants, and autonomous agents. Mem0 runs as a Python or Node library, a self-hosted server, or a managed cloud platform.

68.2kGitHub stars
+459new stars, week of Oct 25
~884.1kdownloads a week, +17% vs 4-week avg
#7in RAG, data and memory this week

Mem0 at a glance

Category: RAG, data and memory
Maintained by: Mem0 (organization, @mem0ai on GitHub)
License: Apache 2.0
Language: Python, TypeScript, Shell
Repository: github.com/mem0ai/mem0
Website: mem0.ai
npm package: mem0ai 3.3.1
PyPI package: mem0ai 2.2.1
Latest release: ts-v3.3.1 on GitHub, Sep 25, 2026
First published: Jun 20, 2023

What Mem0 does

Mem0 retains User, Session, and Agent state so assistants keep context across conversations.
Mem0 extracts memories with a single LLM call that only adds memories, so nothing is overwritten.
Mem0 combines semantic, BM25 keyword, and entity matching in parallel and fuses the scores for retrieval.
Mem0 extracts, embeds, and links entities across memories to improve retrieval.
Mem0 uses time-aware retrieval to rank the right dated memory for questions about current state, past events, or plans.
Mem0 offers a CLI to add and search memories from the terminal, scoped by user ID.
Mem0 provides agent skills that teach coding assistants such as Claude Code, Cursor, and Codex how to integrate Mem0.

How it works

Mem0 runs as a Python or npm library, a self-hosted server started with Docker Compose, or the hosted Mem0 Platform accessed by API key. Mem0 calls an LLM, gpt-5-mini by default, to extract memories and an embedding model, text-embedding-3-small by default, to store them in a vector store. Your app searches memories before generating a reply and adds new messages afterward.

How to install and start using Mem0

With pip

pip install mem0ai

With npm

npm install mem0ai

CLI with npm

npm install -g @mem0/cli

Requirements

Python 3.10 or later, below 4.0, for the Python package.
Node.js 18 or later for the npm package.
An LLM provider; OpenAI is the default, and other LLMs are supported.
A Mem0 API key from app.mem0.ai when using the hosted platform.

First steps

Install the library with pip or npm, or start the self-hosted server with Docker Compose.
Configure an LLM and embedding model; Mem0 defaults to OpenAI's gpt-5-mini and text-embedding-3-small.
Create a Memory instance and call memory.add() with conversation messages and a user_id.
Call memory.search() with a query and user_id filter to retrieve relevant memories before generating a reply.

Who Mem0 is for

Developers building customer support chatbots that need to recall past tickets and user history.
Teams adding persistent, personalized memory to AI assistants and autonomous agents.
Teams that want to run memory on their own infrastructure with a self-hosted server.
Developers using coding agents like Claude Code, OpenCode, or OpenClaw who want memory across sessions.

Limitations to know

Mem0 requires an LLM to function, which adds model calls to each memory add.
Published benchmark scores reflect the managed platform, which includes proprietary optimizations not available in the open-source SDK.
The plain library has no dashboard or auth; the self-hosted server offers only teasers of advanced features.
Hybrid search with BM25 and entity extraction needs the optional NLP extra and a spaCy model download.

Is Mem0 growing? Usage data

In the week of Oct 25–31, 2026, Mem0 ranked #95 on the Devtoolwatch Board and #7 in RAG, data and memory, with a momentum score of 64 out of 100. Its main package had about 884.1k downloads that week, +17% against its own 4-week average.

New GitHub stars per week for the last 12 weeks, from GitHub's stargazer history (today's stargazers by the week they starred).

Show the weekly numbers

New GitHub stars per week for Mem0
Week of	New stars
Aug 9, 2026	538
Aug 16, 2026	527
Aug 23, 2026	513
Aug 30, 2026	461
Sep 6, 2026	477
Sep 13, 2026	498
Sep 20, 2026	396
Sep 27, 2026	529
Oct 4, 2026	470
Oct 11, 2026	442
Oct 18, 2026	420
Oct 25, 2026	459

On Hacker News

Show HN: Mem0 – open-source Memory Layer for AI apps, 201 points and 61 comments (Sep 4, 2024)
Show HN: Mem0 Browser Extension: Shared Memory Across ChatGPT,Claude,Perplexity, 34 points and 4 comments (Nov 4, 2024)
Show HN: OpenMemory – Make your MCP clients more context-aware, 8 points and 0 comments (May 13, 2025)

The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.

Mem0 alternatives

Other RAG, data and memory tools on the Board in the week of Oct 25–31, 2026, by momentum:

RAG, data and memory ranked by momentum, the week of Oct 25–31, 2026
Rank	Tool	Stars	Downloads/wk
#1	XERJXERJ is a local-first search engine and MCP server for AI agents, packaged as one Rust binary with an Elasticsearch-compatible API.	17.9k	–
#2	MemvidMemvid is a single-file memory layer for AI agents that packages data, embeddings, search indexes, and metadata into one portable .mv2 file.	17.3k	–
#3	LEANNLEANN is a Python vector database and RAG library that recomputes embeddings on demand to keep indexes small on personal devices.	13.7k	–
#4	GraphifyGraphify is an AI coding assistant skill that turns a codebase, with its docs, PDFs, images and videos, into a queryable knowledge graph.	140k	480.7k
#5	MarkItDownMarkItDown is a lightweight Python utility from Microsoft that converts PDFs, Office documents, images, audio and other files to…	193.2k	4.7M
#6	FirecrawlFirecrawl is an API and open source project to search, scrape, crawl, and interact with web pages, returning LLM-ready markdown or JSON…	197.4k	2.9M

All RAG, data and memory tools, ranked

Recent Mem0 releases

ts-v3.3.1, Sep 25, 2026
v2.2.1, Sep 25, 2026
deepseek-plugin-v0.3.2, Sep 23, 2026
pi-agent-v0.3.2, Sep 23, 2026
openclaw-v1.2.1, Sep 23, 2026

Mem0 FAQ

Is Mem0 free and open source?

Yes. Mem0's source code is on GitHub (mem0ai/mem0) under the Apache 2.0 license, so you can use it for free. Its website also links to a pricing page (https://mem0.ai/pricing) for any paid plans.

How do I install Mem0?

Run pip install mem0ai. It needs Node.js >=18. The install section above has the other options from the project's README.

Is Mem0 still maintained?

Yes. The mem0ai/mem0 repository was last updated on Oct 21, 2026, and the latest release, ts-v3.3.1 on GitHub, came out on Sep 25, 2026.

How popular is Mem0?

Mem0 has 68.2k GitHub stars as of Nov 1, 2026. It gained 459 new stars in the week of Oct 25–31, 2026, against a 12-week average of 478 a week. Its main package had about 884.1k downloads that week, +17% against its own 4-week average. That made it #7 among RAG, data and memory tools on the Devtoolwatch Board.

Mem0 vs Supermemory: how do they compare?

Mem0 and Supermemory are both tracked by Devtoolwatch as RAG, data and memory tools. Mem0 has 68.2k GitHub stars and Supermemory has 31.5k. In the week of Oct 25–31, 2026, Mem0's main package had 884.1k downloads and Supermemory's had 209.2k. Supermemory is a memory and context engine for AI apps and agents, available as an API, plugins, an MCP server, and a self-hosted binary. Supermemory review.

Mem0 vs Zep: how do they compare?

Mem0 and Zep are both tracked by Devtoolwatch as RAG, data and memory tools. Mem0 has 68.2k GitHub stars and Zep has 5k. In the week of Oct 25–31, 2026, Mem0's main package had 884.1k downloads and Zep's had 259.5k. Zep is a managed context and agent memory platform, Zep Cloud, with SDKs, examples and framework integrations for building AI agents. Zep review.

What are the best alternatives to Mem0?

Other RAG, data and memory tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: XERJ (#1), Memvid (#2), LEANN (#3), Graphify (#4), MarkItDown (#5). The alternatives table above has their numbers.

What is Mem0?

Mem0 is a memory layer for AI assistants and agents. Mem0 remembers user preferences, retains user, session, and agent state, and retrieves relevant memories so conversations stay personalized over time.

Is Mem0 open source?

Yes. Mem0's repository and packages are released under the Apache-2.0 license. The managed platform includes proprietary optimizations not in the open-source SDK.

Can Mem0 be self-hosted or run locally?

Yes. Mem0 runs as a local library or as a self-hosted server started with docker compose up -d, served at localhost:3000. Self-hosted auth is on by default.

How does Mem0 work?

Mem0 uses an LLM to extract memories from messages in a single pass that only adds memories. Mem0 links entities and retrieves with semantic, BM25 keyword, entity, and time-aware signals fused together.

How do I use Mem0 with Claude Code or OpenClaw?

Mem0 offers agent skills installable with npx skills add for Claude Code, Cursor, OpenClaw, and others. Mem0 also publishes plugins for OpenClaw, OpenCode, Pi Agent, and DeepSeek.

Sources

Mem0 README badge

Maintainers can show the tool's weekly Board rank in their README:

[![Devtoolwatch](https://index.devtoolwatch.com/badge/mem0.svg)](https://index.devtoolwatch.com/tools/mem0/)

How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].