Mistral Document & Vision

describeImage

7 credits per call

Analyze and describe an image using AI vision capabilities. Can identify objects, scenes, text, and provide detailed descriptions.

Try it

Open in playground
kirha tools run mistraldocuments_describeImage \
  --input '{
    "imageUrl": "<imageUrl>"
  }'

Uses the key saved by kirha auth login.

  • imageUrlstring, required

    URL of the image to analyze (supports PNG, JPEG, WEBP, GIF, max 10MB)

  • detailLevelstring, default "detailed"

    Level of detail in the description

    One of brief, detailed

  • promptstring, default "Describe this image in detail."

    Custom prompt to guide the image analysis (e.g., 'What objects are in this image?', 'Extract text from this image')