MemOS

MemoryApache-2.0

MemOS is a memory system for language model agents that stores and retrieves memories across tasks. It is published by MemTensor and offered as a cloud API, a self-hosted service, or local plugins for specific agent tools.

The README compares four setups: the MemOS Cloud API, a self-hosted MemOS service started with Docker Compose together with Neo4j and Qdrant, a cloud plugin and a local plugin that stores data in SQLite on the user's machine.

Pinout

  • MCP server: not checked yet
  • HTTP API: supported, verified
  • command line: not checked yet
  • Python SDK: not checked yet
  • TypeScript SDK: not checked yet
  • OpenAPI description: not checked yet
Where each pin comes from
InterfaceStatusSource we readRead on
MCP serverNot checked yet——
HTTP APISupported, verifiedREADME, section "Self-Host the MemOS Service" (MemOS API via docker compose)
Command lineNot checked yet——
Python SDKNot checked yet——
TypeScript SDKNot checked yet——
OpenAPI descriptionNot checked yet——

Facts

Facts about MemOS
PricingCloud plans: Free $0 a month with 50K memory adds and 20K searches; Starter and Pro shown at $0 for now instead of $19 and $286 a month; Enterprise on request. MemOS pricing page, read .
LicenseApache-2.0. GitHub API, repository license field, read .
Main languageTypeScript (as detected by GitHub). GitHub API, repository record, read .
PublisherMemTensor (GitHub organization). GitHub API, repository record, read .
Latest GitHub releasev2.0.36, published 2026-11-09. GitHub API, latest release, read .
Self-hosted stackMemOS API with Neo4j and Qdrant, started by docker compose. README in the repository, read .
Hosted optionMemOS Cloud API. README in the repository, read .
Last commit. GitHub API, latest commit on the default branch of MemTensor/MemOS, read .

Limits

  • The self-hosted service needs a graph database and a vector database running next to it, both named in the setup table of the README.

Strengths

  • The README states where data lives for each of its four setups.

Pricing

PricingCloud plans: Free $0 a month with 50K memory adds and 20K searches; Starter and Pro shown at $0 for now instead of $19 and $286 a month; Enterprise on requestMemOS pricing page, read 16 Nov 2026

As written on MemOS's own page on the read date. Prices change: check there before you pay.

Alternatives to MemOS

All memory tools for agents
  1. Agent memory system with retain, recall and reflect operations.

    Memory
  2. Open-source memory layer storing agent context across sessions and models.

    Memory
  3. Memory layer that wraps LLM clients to store and recall agent state.

    Memory
  4. Stateful agents whose memory and identity persist across sessions and machines.

    Memory

Last check: . Method: public repository and docs read automatically, page and day written down. Next check due within 90 days. How we check

Is this your tool? Write to hello@agent-tools.directory from an address on its domain to correct a pin or a fact.