Hermes

Hermes is Nous Research’s self-improving agent. It accepts a custom OpenAI-compatible model endpoint, so you can point it at Routeplane and route its model calls across the whole registry.

Prerequisites

• Routeplane running — local proxy at http://127.0.0.1:4356. Routeplane Cloud is on the Phase D roadmap and not yet shipping; for now, self-host with your own provider keys.
• Hermes installed. See NousResearch/hermes-agent for setup.

Point Hermes at Routeplane

Set the model provider to custom and the base URL to Routeplane’s /v1 root in ~/.hermes/config.yaml:

model:
  provider: custom
  base_url: http://127.0.0.1:4356/v1
  api_key: local-placeholder    # any value for the local proxy
  model: openai/gpt-4o

The same can be set from the CLI:

hermes config set model.base_url http://127.0.0.1:4356/v1

**No key for the local proxy** — `api_key` can be any placeholder (the loopback proxy ignores it). Routeplane Cloud is on the Phase D roadmap; the cloud endpoint is not yet shipping. Confirm the exact field names against the [Hermes configuration docs](https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md) for your version. **Want to run on Hermes models specifically?** Nous's Hermes models are in the registry too — request `nousresearch/hermes-4-405b` (or any provider-qualified Hermes id) as the model and Routeplane routes to whichever provider serves it.

Pick a model

The model field takes any registry id in provider/model form, optionally with a :cost / :latency variant. See Models.

Learn more

• Hermes — configuration
• Model fallback · Provider selection