Context : With the increasing knowledge of the terrestrial planets due to recent space probes it is possible to model their rotation with increasing accuracy . Despite that fact , an accurate determination of Venus precession and nutation is lacking Aims : Although Venus rotation has been studied in several aspects , a full and precise analytical model of its precession-nutation motion remains to be constructed . We propose to determine this motion with up-to-date physical parameters of the planet Methods : We adopt a theoritical framework already used for a precise precession-nutation model of the Earth , based on a Hamiltonian formulation , canonical equations and an accurate development of the perturbing function due to the Sun . Results : After integrating the disturbing function and applying the canonical equations , we can evaluate the precession constant \dot { \Psi } and the coefficients of nutation , both in longitude and in obliquity . We get \dot { \Psi } = 4474 " .35 / Jcy \pm 66.5 , corresponding to a precession period of 28965.10 \pm 437 years . This result , based on recent estimations of the Venus moment of inertia is significantly different from previous estimations . The largest nutation coefficient in longitude with an argument 2 L _ { S } ( where L _ { S } is the longitude of the Sun ) has a 2 ” 19 amplitude and a 112.35 d period . We show that the coefficients of nutation of Venus due to its triaxiality are of the same order of amplitude as these values due to its dynamical flattening , unlike of the Earth , for which they are negligible . Conclusions : We have constucted a complete theory of the rotation of a rigid body applied to Venus , with up-to-date determinations of its physical and rotational parameters . This allowed us to set up a new and better constrained value of the Venus precession constant and to calculate its nutation coefficients for the first time .