Skip to main content

OpenRouter

Use the OpenRouter connector to send and receive messages via multiple LLMs (e.g., GPT-4, Claude, Mistral, and more). These operations allow seamless integration of AI-powered chat or completion capabilities into your workflows.

Prerequisites​

To use the OpenRouter connector, you need a valid API Key from the OpenRouter Dashboard. Ensure that the selected models are supported and accessible under your account.

For more information, visit the OpenRouter API Documentation.

Supported Authentication Methods​

  • API Key (passed via Authorization request header)

Operations That Can Be Performed Are Listed Below​

Chat Completions​

Send Chat Message​

Send a chat-style message to a selected model (e.g., GPT-4o).

Send Multi Chat Message​

Send a multi-chat-style message to a selected model (e.g., GPT-4o).

Image Analysis​

Send Image for Analysis​

Send an image input to supported multimodal models (e.g., GPT-4o) for analysis, such as understanding content or answering related questions.

Generation Details​

Fetch Generation Details​

Retrieve metadata or results of a previously generated response using its generation ID, useful for tracking or auditing completions.

Completion Parameters​

You can optionally include additional settings to control model behavior:

  • temperature: Controls randomness (default 1.0)
  • max_tokens: Limits the response length

Validation​

To validate your key, make a test request to the /chat/completions endpoint. A successful response confirms your key is valid and active.

Troubleshooting​

If authentication fails:

  • Ensure your API Key is copied correctly.
  • Confirm the key begins with sk-or-.
  • Check that the key has not been revoked or expired.
  • Verify the Authorization header is formatted as: