Hermes Guide

Deploy NousResearch's Hermes agent on Maritime.

Overview

Hermes is a self-improving Python agent from Nous Research with persistent memory and a multi-platform gateway. It ships with a TUI dashboard and a JSON HTTP API.

Docker Image

Image: nousresearch/hermes-agent:latest

Gateway on port 8642, dashboard on port 9119, HOME at /opt/data.

How It Works

Hermes has no built-in openai provider — direct OpenAI access goes through itsopenrouter provider, which is OpenAI-compatible and honorsOPENROUTER_BASE_URL. Maritime points that base URL at the LLM proxy so OpenAI-style env vars work transparently.

# The container runs with:
HERMES_HOME=/opt/data
HERMES_DASHBOARD=1
HERMES_DASHBOARD_HOST=0.0.0.0
HERMES_DASHBOARD_TUI=1
API_SERVER_ENABLED=true
API_SERVER_HOST=0.0.0.0
API_SERVER_KEY=<auto-generated>
GATEWAY_ALLOW_ALL_USERS=true
HERMES_INFERENCE_PROVIDER=openrouter
HERMES_TUI_PROVIDER=openrouter
HERMES_INFERENCE_MODEL=gpt-4o
OPENROUTER_BASE_URL=https://api.maritime.sh/api/llm/v1

Environment Variables

Set OPENAI_API_KEY if you want to bring your own; otherwise Maritime injects a per-agent proxy token. The agent reads the same env var because OpenRouter falls back to OPENAI_API_KEY when no OpenRouter key is set.

Tier Requirements

Hermes loads memory + dashboard at boot, so the Smart tier (512 MB) is tight. Use Extended or higher for stable long-running sessions.

Deploy

Select Template → Hermes Agent in the Create Agent modal, or:

maritime create -n my-agent -t hermes --tier extended && maritime deploy my-agent