Back to docsRecipe

Recipe: Image caption generator

Build an automated pipeline that generates descriptive captions for any image using Meridian's vision model. Ship a production-ready captioning service in under 15 minutes.

What you'll build

  • REST endpoint that accepts an image URL and returns a caption
  • Batch processing for bulk image libraries
  • Style controls: concise, detailed, or alt-text optimized

Prerequisites

  • Meridian API key with vision model access
  • Python 3.10+ or Node.js 18+
  • Basic familiarity with HTTP clients

Quick start

Send a POST request to the Meridian vision endpoint with your image and desired caption style. The model returns a natural-language description optimized for your use case.

POST /v1/vision/caption Authorization: Bearer $MERIDIAN_KEY Content-Type: application/json {"image_url": "...", "style": "detailed"}