In recent years, new algorithms and cryptographic protocols based on the laws of quantum physics have been designed to outperform classical communication and computation. We show that the quantum world also opens up new perspectives in the field of complex networks. Already the simplest model of a classical random network changes dramatically when extended to the quantum case, as we obtain a completely distinct behavior of the critical probabilities at which different subgraphs appear. In particular, in a network of N nodes, any quantum subgraph can be generated by local operations and classical communication if the entanglement between pairs of nodes scales as N −2 .On the one hand, complex networks describe a wide variety of systems in nature and society, as chemical reactions in a cell, the spreading of diseases in populations or communications using the Internet [1]. Their study has traditionally been the territory of graph theory, which initially focused on regular graphs, and was extended to random graphs by the mathematicians Paul Erdős and Alfréd Rényi in a series of seminal papers [2,3,4] in the 1950s and 1960s. With the improvement of computing power and the emergence of large databases, these theoretical models have become increasingly important, and in the past few years new properties which seem universal in real networks have been described, as a small-world [5] or a scale-free [6] behavior.On the other hand, quantum networks are expected to be developed in a near future in order to achieve, for instance, perfectly secure communications [7,8]. These networks are based on the laws of quantum physics and will offer us new opportunities and phenomena as compared to their classical counterpart. Recently it has been shown that quantum phase transitions may occur in the entanglement properties of quantum networks defined on regular lattices, and that the use of joint strategies may be beneficial, for example, for quantum teleportation between nodes [9,10]. In this work we introduce a simple model of complex quantum networks, a new class of systems that exhibit some totally unexpected properties. In fact we obtain a completely different classification of their behavior as compared to what one would expect from their classical counterpart.A classical network is mathematically represented by a graph, which is a pair of sets G = (V, E) where V is a set of N nodes (or vertices) and E is a set of L edges (or links) connecting two nodes. The theory of random graphs, aiming to tackle networks with a complex topology, considers graphs in which each pair of nodes i and j are joined by a link with probability p i,j . The simplest and most studied model is the one where this probability is independent of the nodes, with p i,j = p, and the resulting graph is denoted G N,p . The construction of these graphs can be considered as an evolution process:Evolution process of a classical random graph with N = 10 nodes: starting from isolated nodes, we randomly add edges with increasing probability p, to eventually get the co...