Image optimization

Dynamically generate compressed image transformations via URL parameters, presets or URL-based rules

To load fast on any device around the world, images need to be resized, cropped and compressed. No need to push a 20-megapixel image to deliver a 1170px wide thumbnail on a smartphone. Filerobot will manipulate and transform your images dynamically and optimize for web delivery, thus saving you days in image management.

Filerobot's image manipulation and transformation features rely on Cloudimage, a standalone media optimization CDN developed by Scaleflex.

If you do not need storage for your origin (master) images and Digital Asset Management but just on-the-fly image and video transformation + CDN delivery, check-out Cloudimage. Cloudimage is a media proxy pulling your media assets from your origin hosting (web server, S3 bucket, etc.), caching, compressing and delivering them over CDN.

We have documented the most common resizing functions here. If you have any requests for undocumented features, contact us and we will do our best to help.

The basics

Everything starts with the basic URL format of a Filerobot URL:



Your Filerobot token


Here, you can use the full path to the image in your Filerobot container, e.g. demo/part2/image.jpg


Image manipulation operations like width, height, crop, fit, rotate, and trim. For a full list, see Operations.


Use filters like contrast, pixellate, blur, grayscale, and sharpen to apply different effects to your image. All filters can be found in the Filters.


Place static or dynamic watermarks on your images, Watermarking.

The hostname can be replaced by a custom CNAME. Our Support team can set it up for you.

Filerobot automatically compresses these images into the most optimal delivery format (WebP, AVIF, ...) based on the user's device and context and accelerates them over multiple Content Delivery Networks from tier 1 CDN providers such as Akamai, Fastly, AWS Cloudfront and CDNetworks.



Resize operation: /docs/paris.jpg?width=400

Resize and convert to grayscale: /docs/paris.jpg?width=400&gray=1

Last updated