Configure Hermes

Hermes Agent connects to InOneAPI through a custom OpenAI-compatible provider. Saving the endpoint, credentials and model in the official model wizard is recommended.

  1. Run hermes model and choose Custom endpoint.
  2. Set the name to InOneAPI and the Base URL to https://api.inoneapi.com/v1.
  3. Enter your InOneAPI API key and select Chat Completions as the API mode.
  4. Select YOUR_MODEL_ID from the model detail page; if automatic discovery does not list it, enter the ID manually.
  5. Save and select that model. In an existing session you can switch with /model. On Windows, WSL2 is recommended.

The wizard stores the credentials and provider in the layout used by your current version, and the model detail page no longer generates a Hermes configuration file. Manual configuration is also supported, but different versions may use model, the older custom_providers list, or the newer providers dictionary, and these structures cannot be mixed directly. Do not treat the CC Switch provider fragment as a replacement for the whole config.yaml.

Verify the configuration

After selecting the model, run a read-only task, then confirm the public model ID, status code, tokens, cost and trace in the console. Next, run a simple task that requires tool calling to confirm that the selected model supports the tool format used by Hermes.

Common issues

  • YAML fails to parse: indent with spaces and do not mix in tabs.
  • 401: rerun the wizard and check the API key; if you set environment variables manually, check whether an old value overrides the credential.
  • 404: check the model ID and the /v1 in base_url.
  • Tool calling fails: a working protocol does not mean the model has tool capabilities. Switch models or disable the affected tools.

Create a dedicated key for Hermes and set a monthly budget and model restrictions, so that automated tasks can be identified separately in team usage.

Optional: import into CC Switch

On the model detail page, select this tool and an API key, then click “Open CC Switch and import”. Use the official CC Switch release; the import link already contains the current model ID, so you do not need to enter it again. After confirming the import, finish the connection by enabling the provider and setting the default model for this tool. If it reports that this app is unsupported, update CC Switch or follow the manual steps in this document.

The import link only adds a provider. It does not mean the model is already the default or that a request succeeded. The link contains an API key, so do not share it. Use either the manual setup or the CC Switch import, not both.

The current CC Switch source generates base_url, api_key, api_mode: chat_completions and models for Hermes in snake_case. After importing, you still select the provider and model with hermes model or /model.

Official reference: Hermes AI Providers.

Hermes · Documentation · InOneAPI