Zihin icon

Zihin

Process media with Zihin AI - Image Analysis, Audio Transcription, Document Parsing

Actions2

Image → Analyze

AI-generated

Summary

Analyze an image with Zihin Vision AI, optionally following a prompt, extracting text with OCR, and selecting a detail level or provider. Provide an image URL or base64 data URI; the node sends each input item to the image analysis API and returns the analysis data with execution metadata.

Inputs

  • Image (required) — Image URL or base64 data URI to analyze.
  • Prompt — Instructions for the analysis. If empty, the API uses its default description.
  • Additional Options — Extract Text (OCR) enables text extraction; Detail Level can be auto, low, or high; Provider can be auto, OpenAI (GPT-4o), or Anthropic (Claude). The default auto values for detail and provider are omitted from the request.
  • Options — Base URL Override changes the API endpoint. Non-Zihin domains require Allow Custom Base URL; localhost requires Allow Localhost. Timeout is in milliseconds; 0 uses the operation default, and positive values are capped at 900,000 ms. Requires Zihin API credentials.

Output shape

One n8n item per input item, containing the analysis response fields plus a _meta object.

_meta contains success and execution_time_ms. If Continue On Fail is enabled, failed items instead contain error and success: false; otherwise the operation throws an error.

Examples

Example 1: Extract visible text from a product photo

Set Image to the photo URL and enable Extract Text (OCR). Leave Prompt empty to use the default analysis.

Links

Discussion