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

Synapse

Synapse is an open-source Python library that provides self-hosted temporal knowledge-graph memory for AI agents. It uses Graphiti and FalkorDB, with a hippocampus layer that scores memory importance, manages forgetting, detects novelty and contradictions, and expands retrieval with related graph context. The repository documents integration as a Hermes Agent memory provider.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Synapse — 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 Synapse when an agent needs to recall relationships and changes across conversations, and you can provide a self-hosted FalkorDB instance plus an OpenAI-compatible LLM endpoint.

Who it is for

  • Developers building AI agents that need persistent memory across conversations.
  • Hermes Agent users who want graph-based temporal memory, either in place of or alongside native memory.
  • Teams that want to host the memory database themselves and choose an OpenAI-compatible LLM endpoint.

Documented uses

  • Retain connected facts and retrieve what was true at an earlier point in time.
  • Score memories by importance and let lower-priority information decay.
  • Detect novelty or contradictions as new episodes are ingested.
  • Expand a search result with its surrounding knowledge-graph context.
Before you choose

Requirements & limits

The repository specifies an MIT license. The documented setup uses Python 3.11 or later and a self-hosted FalkorDB instance.

  • Synapse is documented as a Hermes Agent memory provider, so its primary integration path is tied to Hermes.
  • The README describes both a brain mode that replaces native memory and a supplementary mode that retains it and mirrors native writes to the graph.
  • LLM endpoints must be OpenAI-compatible; the configuration lists an API key and base URL as required variables.
  • The README presents performance numbers as projected estimates based on architectural analysis, not measured guarantees.

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 Synapse as a Hermes memory provider

Install Synapse, start FalkorDB and configure Hermes to use Synapse for agent memory.

Before you start

  • Python 3.11 or later
  • Docker for the documented FalkorDB container setup
  • A Hermes Agent installation
  • An OpenAI-compatible LLM endpoint, API key and base URL

Steps

  1. Start FalkorDB with the documented Docker command, publishing port 6379.

  2. Install Synapse from its Git repository with pip.

  3. Set Hermes's memory provider to synapse.

  4. Configure SYNAPSE_FALKORDB_HOST, SYNAPSE_LLM_API_KEY and SYNAPSE_LLM_BASE_URL in the Hermes environment file.

Expected outcome

Hermes is configured to use Synapse with a FalkorDB host and an LLM endpoint.

Guide limitations

  • This recipe covers the README's configuration stage; it does not independently verify a running setup.
  • The repository documents Hermes integration rather than a standalone user-facing assistant.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
Ardha-Eco-System/synapse · Official homepage title · retained in its source language
How we review sourcesSuggest a correction