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

Memfuse

MemFuse Core Services is a Python server that gives LLM applications persistent, queryable memory across conversations and sessions, using a database-backed memory layer and a separate client SDK for application integration.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Memfuse — 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 MemFuse when an LLM application needs durable, queryable conversation memory beyond the current prompt or session.

Who it is for

  • Developers building LLM applications that need to retain user context between sessions.
  • Teams seeking a self-hosted memory service to connect to an LLM application.

Documented uses

  • Retain user preferences and conversational context across sessions.
  • Retrieve earlier facts or events when responding to later queries.
  • Reduce repeated transmission of chat history to an LLM.
Before you choose

Requirements & limits

The project is open source under the Apache 2.0 license.

  • The repository is for core server services; application integration uses the separate MemFuse Python SDK.
  • Local setup uses Docker and TimescaleDB, with pgvector and a built-in pgai-like embedding implementation.
  • The README describes combined vector, graph and keyword search, layered memory, and multi-tenant scoping.
  • The repository lists LangChain, AutoGen, Vercel AI SDK and direct LLM calls among integration options.

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

Set up the MemFuse core server locally

Install the repository dependencies and start the server using its documented local setup.

Before you start

  • Docker and Docker Compose installed
  • A running Docker daemon
  • The TimescaleDB image timescale/timescaledb-ha:pg17
  • Python project dependencies installed with Poetry or pip

Steps

  1. Clone https://github.com/memfuse/memfuse and change into the repository directory.

  2. Check Docker availability and pull the TimescaleDB image with docker pull timescale/timescaledb-ha:pg17.

  3. Install dependencies with poetry install, or with pip install -e .

  4. Start the server with poetry run python scripts/memfuse_launcher.py, or python scripts/memfuse_launcher.py when using pip.

Expected outcome

The README’s documented local server setup is installed and launched.

Guide limitations

  • The guide covers server setup; application use requires the separate MemFuse Python SDK.
  • The repository notes that Docker must be running for database startup.
Documentation 1 ↗
Evidence and transparency

Sources & review

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