OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Mindmemos

MindMemOS is an open-source persistent memory service for AI agents that stores and retrieves user and task context, connects file-based knowledge, and supports distilling skills from interactions.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Mindmemos — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use MindMemOS when an agent workflow needs persistent context across interactions or agents, or when developers want a shared memory service accessible through an API, SDK or plugin.

Who it is for

  • Developers connecting AI agents to reusable user, project or task context.
  • Teams integrating memory through an HTTP API, Python SDK or CLI, or supported agent plugins.
  • Operators seeking a locally self-hosted memory service.

Documented uses

  • Preserve user profiles, preferences, project facts and tool experience for reuse across agents.
  • Retrieve relevant memories for an agent interaction and write conversations back to memory.
  • Turn experience memories into skill candidates and incorporate execution results, failure traces and feedback into memory.
  • Connect agent memory with file-based knowledge systems.
Before you choose

Requirements & limits

The README documents an official cloud service that requires requesting an API key, as well as local self-hosting with locally configured keys. It lists HTTP API, Python SDK/CLI, OpenClaw plugin and DeepSeek Harness plugin access methods.

  • Local setup requires chat and embedding model routers; reranking is optional.
  • Embedding dimensions must match the Qdrant dimension configuration.
  • The local stack uses Docker-managed dependencies and FastAPI, with Qdrant, Neo4j and Kafka listed as core dependencies.
  • The DeepSeek Harness plugin is described as automatically recalling and writing MindMemOS memories.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare a local MindMemOS deployment

Create the configuration files and configure the model routers and API key needed before starting the local service.

Before you start

  • The MindMemOS repository and its deployment instructions.
  • uv for dependency management and local commands.
  • A chat model router and an embedding model router configured in config/mindmemos/dev.yaml; reranking is optional.
  • Embedding dimensions that match the Qdrant dimension configuration.

Steps

  1. Copy .env.example to .env.

  2. Copy config/mindmemos/dev.example.yaml to config/mindmemos/dev.yaml.

  3. Configure the required model routers in config/mindmemos/dev.yaml.

  4. Set an API key and its bound project_id in config/mindmemos/api_keys.yaml.

Expected outcome

Configuration files are prepared for the subsequent local service startup stage.

Guide limitations

  • These steps cover preparation only, not starting or verifying the service.
  • The reranking model router is optional; the chat and embedding model routers are required.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
mindscale-noah/MindMemOS · Official homepage title · retained in its source language
How we review sourcesSuggest a correction