Aurora AIInstall

Summary

Analyze an image by sending its binary data and an optional descriptive prompt to the Aurora AI multimodal API, which returns a text-based analysis or description of the image.

Inputs

• System Instruction — Optional instructions to define the assistant's behavior or personality during analysis.
• Temperature — Number between 0 and 2 controlling randomness in the AI's output (default 0.5).
• Top P — Number between 0 and 1 controlling nucleus sampling probability mass for the output (default 0.9).
• Max Output Tokens — Maximum number of tokens allowed in the API's textual response (default 1024).
• Prompt — Text instruction for how the image should be analyzed or described (default 'Describe this image.').
• Binary Property (required) — Name of the binary property in the incoming item that contains the image to analyze (default 'data').

Output shape

A single JSON object containing the API response with the image analysis text result.

The operation expects image binary data under the specified binary property in the input item. If the binary data is missing, the node throws an error unless 'Continue On Fail' is enabled. The response JSON contains the AI-generated description or analysis, and tokens limits and generation parameters can be configured via inputs.