Rahyana SimpleInstall

Summary

Send chat messages to a specified Rahyana AI model and receive generated responses, optionally using and updating conversational memory and streaming output.

Inputs

• Model (required) — The model slug identifying the specific Rahyana AI model to use (e.g., 'google/gemini-2.0-flash-lite-001').
• Messages (required) — An array of chat messages to send to the AI, formatted as [{role, content}].
• System Prompt — A system message prepended to set the AI's behavior or context for the conversation.
• Options — Additional parameters controlling the AI response such as temperature (creativity), maximum tokens, streaming response, callback URLs for streamed data, whether to return raw API output, using input memory field for conversational context, writing updated memory to output, request timeout, and retry count for transient errors.

Output shape

A single item with JSON containing either the generated chat completion content and optionally updated memory field or raw API response if requested.

If streaming is enabled, the node sends streamed chat events and optionally forwards each event to a callback URL, outputting a single summary object indicating streaming status. Without streaming, the node outputs one JSON containing the AI response content. If memory is used, input memory data can be merged into messages, and updated memory is written to a specified output field. The node handles retries on HTTP 429 or server errors with exponential backoff and supports request timeout via AbortController.