Judging by the downloaded collection of new files in the official Google profile on GitHub, the search giant is working on a new image format called Pik. According to the description of the project in the readme file, it is a technology of compression with loss of quality, created for use on the Internet. At the moment this is practically all that is known.

For those who follow Google's efforts in the development of image and video compression formats, this description may recall the open WebP standard, which was first introduced in September 2010 and is now used in a number of products. At that time, the company, like in the case of Pik, promoted it as a new standard for compressing web graphics with loss of quality, seeking to replace outdated and inadequate JPEG and GIF.

It is not yet clear what exactly Pik will differ from WebP (and even more from other modern formats of web images): Google only mentions that the new format requires a processor with support for AVX2 and FMA instructions, for example, Haswell. One can only assume that Pik is based on a completely new algorithm that provides significant superiority over WebP and JPEG in terms of the ratio of file size and quality.

According to the description, Pik is not an official Google product, although it is developed by the search giant. Apparently, this means that while Google is not going to make an official announcement of this product, let alone promote it as a real alternative.

Perhaps, as the development of Pik will be able to become a full-fledged format for storing images, but so far this is only one of many experimental projects of the company.

The project page on GitHub offers interested developers access to the encoder and decoder Pik, so that interested specialists can already evaluate the advantages and disadvantages of the format. Of course, if they are not afraid of the words that Pik is at an early stage of research and is not suitable for use in any full-fledged tasks.