Hermes Agent

Set up with the wizard

Hermes Agent works with any OpenAI-compatible API. Run its model wizard from your terminal (not inside a chat session) and choose a custom endpoint:

hermes model
# → Custom endpoint (self-hosted / VLLM / etc.)
# → API base URL:  https://api.greentokens.io/v1
# → API key:       sk-gt-…
# → Model name:    claude-sonnet-5

The wizard saves your answers to Hermes's config file, so they survive restarts.

Or edit the config

The same setup by hand. Keep the key in ~/.hermes/.env and reference it with key_env, so it never sits in the config file:

model:
  provider: custom
  default: claude-sonnet-5
  base_url: https://api.greentokens.io/v1
  key_env: GREENTOKENS_API_KEY

• base_url includes /v1.
• default is any text model ID from the models page.

Switching models

Inside a session, switch to another GreenTokens model with /model custom:model-id, for example /model custom:gpt-6-sol.