Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.fluximage.io/llms.txt

Use this file to discover all available pages before exploring further.

Fluximage can write alt text for your images automatically. When you enable this feature, each image is analyzed by GPT-4o mini (OpenAI Vision), which reads the visual content and produces a concise, descriptive alt text string ready to paste into your HTML or CMS.

​
How it works

After your images are compressed, Fluximage sends each one to the OpenAI Vision API. The model looks at the image and generates alt text that describes what it sees, taking into account any context or keywords you provide. The result is returned alongside the optimized file in your download. Each generated alt text is:
  • Up to 80 characters — short enough for screen readers and search engines
  • Context-aware — shaped by the keywords and context you supply
  • In the language you select — output matches your chosen language, not the language of the original filename

​
Supported languages

English, Spanish, Catalan, French, German, Italian, Portuguese, Dutch, Polish, Russian, Chinese, Japanese, Korean, Arabic, Hindi

​
Providing keywords and context

You can steer the output by filling in two optional fields before processing:
  • Keywords — specific terms you want the alt text to incorporate (e.g. running shoes men red)
  • Context — a short description of your site or use case (e.g. ecommerce store selling sportswear)

​
Example

InputValue
ContextOnline store selling athletic footwear
Keywordsred running shoes men
Output alt textRed men’s running shoes with white sole on white background
Without context or keywords, the model describes only what it sees. With them, the output is more likely to match the terminology your customers search for.

​
Plan requirement

AI alt text requires a Starter plan or higher. It is not available on the free plan.
Each image processed with AI alt text enabled uses 1 AI credit. If you also enable SEO rename for the same batch, both features together still count as 1 credit per image — not 2.