We study a d-dimensional wave equation model (2 ≤ d ≤ 4) with quadratic non-linearity and stochastic forcing given by a space-time fractional noise. Two different regimes are exhibited, depending on the Hurst parameter H = (H 0 , . . . ,, the model must be interpreted in the Wick sense, through a renormalization procedure.Our arguments essentially rely on a fractional extension of the considerations of [12] for the twodimensional white-noise situation, and more generally follow a series of investigations related to stochastic wave models with polynomial perturbation.Since the pioneering works of Mandelbrot and Van Ness, fractional noises have been considered as very natural stochastic perturbation models, that offer more flexibility than classical white-noise-driven equations. The involvement of fractional inputs first occured in the setting of standard differential equations and, even in this simple context, is known to raise numerous difficulties due to the nonmartingale nature of the process. Sophisticated alternatives to Ito theory must then come into the picture, whether fractional calculus, Malliavin calculus or rough paths theory, to mention just the most standard methods.More recently, fractional (multiparameter) noises have also appeared within SPDE models. A first widely-used example is given by white-in-time colored-in-space Gaussian noises, that can be treated in the classical framework of Walsh's martingale-measure theory [25], or with Da Prato-Zabczyk's infinitedimensional approach to stochastic calculus [6]. Such noise models have thus been applied to a large class