Skip to Content
Model API ReferenceAuthentication headers

Authentication headers

Create a key in the Fenno console  and keep it server-side. A key’s existence does not guarantee access to every model or unlimited usage. For setup steps, see the authentication guide.

ProtocolAuthenticationAdditional JSON headers
OpenAI-compatibleAuthorization: Bearer <FENNO_API_KEY>Content-Type: application/json
Anthropic Messagesx-api-key: <FENNO_API_KEY>anthropic-version: 2023-06-01, Content-Type: application/json
Geminix-goog-api-key: <FENNO_API_KEY>Content-Type: application/json

The Anthropic SDK adds its normal version headers. Do not configure conflicting api_key and auth_token settings for the same request. Gemini can use a key query parameter where accepted, but headers reduce the risk of keys leaking through URLs and access logs.

Environment variables

The guides use FENNO_API_KEY as an application-owned variable and pass it explicitly into SDK clients. SDKs do not all read this name automatically:

import os from openai import OpenAI client = OpenAI( base_url="https://api.fenno.ai/v1", api_key=os.environ["FENNO_API_KEY"], )

Tool-specific variables are documented in each integration guide. Restart an application when necessary after changing its environment; a GUI launched before a shell export may not inherit it.

Credential failures

  • 401: check the value actually used, deleted or disabled keys, and conflicting client authentication.
  • 403: check model access and policies instead of repeatedly rotating otherwise valid keys.
  • Redact keys in screenshots, logs and support requests. Never commit them into repository settings or frontend bundles.
  • Rotate a leaked key through the account controls available to you, then update dependent services.

See the error reference and base URL overview.