All Guides

Integration Guide

Hermes Agent + OrcaRouter: Connect the Routed API

OrcaRouter presents multiple upstream models through one OpenAI-compatible endpoint. Hermes Agent can consume that endpoint as a custom provider while OrcaRouter handles the selected route.

How the route works

Hermes Agent → https://api.orcarouter.ai/v1 → OrcaRouter → selected provider/model

Hermes sends the standard chat-completions request. OrcaRouter authenticates the request, maps the requested model to a route, and returns an OpenAI-compatible response. Availability, logging, prices, and provider policies come from OrcaRouter and its upstreams.

Get the three required values

  • Base URL: https://api.orcarouter.ai/v1.
  • API key: create one in your own OrcaRouter account.
  • Model ID: copy it from the current OrcaRouter catalog or models response; do not guess the spelling.

Configure Hermes

# /opt/data/config.yaml
model:
  provider: custom
  base_url: https://api.orcarouter.ai/v1
  api_key: YOUR_ORCAROUTER_KEY
  default: EXACT_MODEL_ID

Restart Hermes after changing the file. If you later switch back to a built-in provider, remove the stale custom base_url and api_key so requests do not continue to route through OrcaRouter.

Use it on a managed instance

On OpenClaw Launch, save the OrcaRouter base URL and key in the Custom API Provider card. Set the instance model to the exact OrcaRouter ID after the key saves. Secrets are configured per account; never paste an operator or platform-owned key into a shared image.

Security and reliability checklist

  • Create a dedicated key for the agent so it can be revoked independently.
  • Set account spending limits or alerts when the router supports them.
  • Confirm whether prompts are logged before sending private data.
  • Test tool calling on the exact route; OpenAI compatibility alone does not guarantee every model follows tool schemas well.
  • Keep a direct or alternate provider plan for outages if the agent is business-critical.

Troubleshooting

  • 401: the OrcaRouter key is missing, revoked, or copied with whitespace.
  • 404: the base URL or model ID is wrong; do not append /chat/completions to the stored base URL.
  • Unknown model: copy the current router ID instead of reusing an OpenRouter ID.
  • Tool loop: select a tool-capable model and verify OrcaRouter preserves tool calls end to end.

Hermes and OrcaRouter FAQ

Can Hermes Agent use OrcaRouter?

Yes. OrcaRouter exposes an OpenAI-compatible API, so Hermes can use it through the custom provider slot.

What base URL should I use?

Use https://api.orcarouter.ai/v1, unless OrcaRouter’s current documentation assigns a different endpoint to your account.

Is OrcaRouter the same as OpenRouter?

No. They are separate routing services with different accounts, keys, model catalogs, policies, and availability.

Related provider guides

Run Hermes Agent around the clock

Deploy Hermes, then attach your own OrcaRouter endpoint and key through the custom provider settings.

Deploy Hermes Agent