The main goal of the presentation is to review the state of art in a field of molecular information processing devices.

There are two basic lines of developing these means.

First of these include attempts to elaborate molecular scale primitive elements such as molecular rectifiers, molecular switches and so on. The most remarkable examples in this field are protein-based associative processors and volumetric memories.

The second is based on the conception of distributed molecular (chemical) information processing media.

Chemical reaction-diffusion media represent information processing means capable to solve efficiently problems of high computational complexity. Distributed character and complex nonlinear dynamics of chemical reactions inherent in the medium is the basis of high parallelism and complex logical operations performed by the medium as primitives. Photo-sensitive catalysts of the medium reactions enable to perform easily input of initial data and output of computational results and to control the dynamics (modes of functioning) of the medium.

It was found during the last decade that chemical reaction-diffusion media can be used effectively for image processing, finding the shortest paths in a labyrinth and solving some other problems of high computational complexity.

Detailed analysis showed that designing media having multilevel architecture and suitable for performing evolutionary computations seemed to be promising way for the further development of reaction-diffusion based devices.