This paper studies the cooperative source seeking problem via a networked multi-vehicle system. In contrast to the existing literature, each vehicle is controlled to the position that maximizes aggregated multiple unknown scalar fields and each sensorenabled vehicle only takes samples of measurements of one scalar field. Thus, a single vehicle is unable to localize the source and has to cooperate with its neighboring vehicles. By jointly exploiting the ideas of the consensus algorithm and the stochastic extremum seeking (ES), this paper proposes novel distributed stochastic ES controllers, which are gradient-free and do not need vehicles' positions, such that the multi-vehicle system of both single integrators and nonholonomic unicycles simultaneously approaches the position of interest. The effectiveness of the proposed controllers is proved for quadratic scalar fields by using the stochastic averaging theory. Finally, illustrative examples are included to validate our theoretical results.