Optimize and control image quality!

Image API for high-volume processing, capable of applying AI algorithms to images as well as conventional image processing algorithms.


JPG 27 kB
Quality 0.92

PNG 141 kB

Click and move the blue slider to compare the 80% compression effects.
Left: Original PNG image of size 141 kB. Right: Optimized image of size 27 kB, processed using quality algorithm fsim-c, quality threshold 0.92, and output format JPEG.


Integrate Anywhere

Use our SDKs to set up and use Pixellena in your technology stack. Browse our tutorials to learn how to easily put Pixellena to use in your projects.

How: To optimize images all you need is a valid API token. Check out the tutorial in three steps to get it up and running now! Read more »

Try the online demo

Upload or provide an image URL and see the results!

How: The online demo can be used to test Pixellena's image optimization API Pixellena. Try the demo »



Task engine to process images in the cloud at scale

Pixellena is built on the task engine Tilemore, that runs in any cloud and uses spot instance capabilities in a convenient way for intensive media-processing tasks.

How: Add your image algorithm and use an API endpoint for Toilmore deployed in specialized infrastructure for you! Read more »

About the operations

Powerful image processing at your fingertips with a simple API


Control image quality

Pixellena allows you to perform image format conversion and optimization, while preserving image quality

How: For lossy images, it does line-search to find the encoded image that best approximates a given quality metric and threshold. Read more »

Focal Point Cropping

Focal point cropping gives you the ability to choose and fine-tune the point of interest of your image for better art direction.

How: Either provide coordinates for a focal point, or simply use one of the algorithms that we have to automatically detect the focal point. Read more »




Produce up-scaled versions of images starting from a lower resolution image which is the best you have available.

How: We use Super Resolution GAN (SRGAN) as the enhancement algorithm. The motive of this architecture is to recover finer textures from the image when upscaling so that the quality cannot be compromised. Read more »

Three top-level stages

Pixellena’s main use case is to prepare an image for visualization in a browser. Basically, after the image is submitted, it passes through three top-level stages:

Step 1: Master

Create a pseudo-master from the ad-hoc input.

Learn more »

Step 2: Shifter

Re-frame and re-scale the master image.

Learn more »

Step 3: Encoder

Convert to WEBP, JPEG, PNG, JPEG2000, or AVIF using encoding tools and quality parameters

Learn more »