ReMe

What is ReMe?

ReMe is a local-first memory system for AI agents from the AgentScope team. It turns conversations and resources into readable, editable Markdown files with frontmatter and wikilinks, so the user keeps the durable memory and can inspect, move or back it up with ordinary tools. Retrieval combines BM25, optional embeddings and wikilink expansion to return line-level passages without loading the whole knowledge base into the agent context. A single memory workspace can be shared by several agents through a CLI, an HTTP service, MCP, a Python API or a SKILL.md, with plugins for Claude Code, Codex, OpenClaw, Hermes Agent and DeepSeek Harness. ReMe Studio is a local web workspace for browsing the files, viewing the memory graph and chatting with a read-only agent. ReMe installs with pip, needs Python 3.11 or newer, and is released under the Apache-2.0 license, so there is no pricing to pay.

Use cases

• Give agents long-term memory across sessions
• Share one memory workspace between several agents
• Search past notes with hybrid retrieval
• Browse and edit agent memory in Studio
• Add memory to Claude Code or Codex
• Consolidate daily notes into lasting knowledge

Who it's for

• AI agent developers
• Python developers
• Coding-agent users
• Researchers

Why use ReMe

• Memory stays as plain Markdown that you own, so you can read, edit and back it up with familiar tools.
• One workspace serves many agents, which avoids tying storage to a single model or host.
• Search returns line-level passages, so agents do not need the whole knowledge base in context.
• Apache-2.0 licensing and a pip install keep adoption simple for Python teams.

ReMe features

• Memory stored as Markdown files with frontmatter and wikilinks
• Auto memory from conversations and resources
• Hybrid search with BM25, optional embeddings and wikilink expansion
• Linked memory graph and file-native entity tags
• CLI, HTTP service, MCP, Python API and SKILL.md interfaces
• Plugins for Claude Code, Codex, OpenClaw, Hermes Agent and DeepSeek Harness
• ReMe Studio local web workspace with an in-browser demo
• Plugin system for extending ReMe with components, steps and jobs
• Docker support

ReMe integrations

• Claude Code
• Codex
• OpenClaw
• Hermes Agent
• DeepSeek Harness
• QwenPaw
• Zvec
• FAISS

How to get started with ReMe

1. Install ReMe with pip install reme-ai[core] on Python 3.11 or newer, or use Docker.
2. Start the service with reme start, which listens on 127.0.0.1:2333 by default.
3. Connect your agent through its plugin, MCP, the CLI, HTTP or the reme_memory skill.
4. Let the agent search memory before answering and pass conversations to auto memory afterward.
5. Open ReMe Studio to browse, edit and explore the memory files and their graph.

ReMe limitations

• Requires Python 3.11 or newer
• You run and maintain the local service yourself
• Building Studio from source needs Node.js 22.13 or newer
• The browser demo of Studio runs no model or background tasks

ReMe pricing

ReMe is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

ReMe alternatives

The main ReMe alternatives listed on Debuly are Mem0, Letta, Zep, Cognee, Supermemory and agentmemory. The table compares pricing and platforms; order is A–Z.

1. 

   agentmemory

   Open-source memory server that lets AI coding agents like Claude Code and Cursor recall past sessions, with local-first search.

   Developer platformsCodingProductivity
   Open sourceVisit agentmemory (official site)
2. 

   Cognee

   Open-source memory engine for AI agents that combines vector search, knowledge graphs and relational storage, with a hosted cloud.

   Developer platformsAutomation and agentsResearch and search
   FreemiumVisit Cognee (official site)
3. 

   Letta

   Platform, formerly MemGPT, for building stateful AI agents with persistent memory, from a terminal app, SDK or API.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Letta (official site)
4. 

   Mem0

   Memory layer for AI agents and apps, available as an open-source package or a managed platform that keeps user context across sessions.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Mem0 (official site)
5. 

   Supermemory

   Memory and context API for AI agents with user profiles, a memory graph, retrieval and connectors, billed by usage.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Supermemory (official site)
6. 

   Zep

   Managed memory and context platform for AI agents, built on the Graphiti knowledge graph, with Python, TypeScript and Go SDKs.

   Developer platformsAutomation and agents
   FreemiumVisit Zep (official site)

ReMe FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from reme.agentscope.io · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.