Aurora AIInstall

Summary

Send retrieved context and a question to the Aurora AI API to generate a response that answers the question based on the provided context.

Inputs

• System Instruction — Optional instructions that define the assistant behavior or personality.
• Temperature — Numeric control for randomness of the model output, ranging from 0 to 2.
• Top P — Numeric control for nucleus sampling to determine probability mass of tokens considered, ranging from 0 to 1.
• Max Output Tokens — Maximum number of tokens allowed in the model response.
• Context Chunks (required) — Retrieved context chunks to inject into the prompt for answering the question.
• Question (required) — The question to be answered using the provided context.

Output shape

a JSON object for each input item containing the model's completion response in the 'json' field, paired with the input item index

The output contains one result per input item. On error, if 'Continue On Fail' is enabled, the output JSON will contain an 'error' message instead of a valid response.