Skip to main content
Install the Dari CLI, then launch Hermes Agent through your organization’s default router:
Hermes runs its chats, messaging platforms, and scheduled tasks through a shared gateway that reads config.yaml, so a single launch cannot route only itself. dari hermes writes Dari into the config that hermes config path reports (it follows HERMES_HOME and the active profile), then starts the installed hermes executable with every following argument. Your other providers, platforms, and settings are kept. New sessions use Dari; run hermes gateway restart to move messaging sessions that are already open. Setup adds a dari entry under providers that offers your router’s models and makes it the main model: model.provider: dari and model.default: dari/routing (or your router’s first model, without Autorouting). It removes the main model’s own base_url, api_key, api_mode, and context_length, which would otherwise steer Dari’s route, and saves what it replaced in dari-setup.json beside config.yaml. To set up without launching, or to undo it and restore your previous main model:
On first use, Dari signs you in if needed and creates and privately caches a Routing key, which setup writes as the entry’s api_key. Gateway includes personal subscriptions, Dari managed billing, and bring-your-own provider keys; the $20/month individual Autorouting plan unlocks the dari/routing router. Gateway serves any model enabled on your router: switch between them with /model custom:dari:dari/model/<model>. To subscribe, run dari hermes --connect and choose Get Autorouting (or press a in the setup model picker). On first launch, setup opens an Accounts page where you can connect one or more ChatGPT subscriptions, connect an API key (Vercel AI Gateway, OpenRouter, or direct provider keys), or choose Dari managed billing. Run dari hermes --connect to open it again. Each launch refreshes the entry, so run dari hermes or dari hermes --setup again after changing the router’s models. Hermes sends Chat Completions requests through the router at /v1. Setup configures a one-million-token context window and sends Hermes’s session id in x-session-affinity, which groups all requests from one Hermes session under the same conversation in Activity.

​
Manual Configuration

If you do not want to use the launcher, create a Routing API key and add this to ~/.hermes/config.yaml:
To use a different model set, edit the default router or Create A Router. To launch dari hermes on that router, add Hermes Agent to its card on the Routers page or run dari router assign-agent <router_id> hermes; the X on the card moves it back to the default router. With manual configuration instead, point base_url at https://routing.dari.dev/{router_id}/v1.