We consider Nash equilibrium problems with mixed-integer variables in which each player solves a mixed-integer optimization problem parameterized in the rivals’ strategies. We distinguish between standard Nash equilibrium problems (NEP), where the parameterization acts only on the players’ cost functions and generalized Nash equilibrium problems (GNEPs), where, additionally, the strategy spaces of the players may depend on the rivals’ strategies. We introduce a branch-and-cut (B&C) algorithm for such mixed-integer games that, upon termination, either computes a pure Nash equilibrium or decides their non-existence. The main idea is to reformulate the equilibrium problem as a suitable bilevel problem based on the Nikaido–Isoda function of the game. We then use bilevel-optimization techniques to get a computationally tractable relaxation of this reformulation and embed it into a B&C framework. We derive sufficient conditions for the existence of suitable cuts and finite termination of our method depending on the setting. For GNEPs, we adapt the idea of intersection cuts (ICs) from bilevel optimization and mixed-integer linear optimization. We can guarantee the existence of such cuts under suitable assumptions, which are, in particular, satisfied for pure-integer GNEPs with decoupled concave objectives and linear coupling constraints. Since ICs may become arbitrarily shallow, we complement them by a fallback strategy that fixes recurring integer profiles and ensures finite termination for a relevant settings. For NEPs, we show that suitable cuts always exist via best-response inequalities and prove that our B&C method terminates in finite time whenever the set of best-response sets is finite. We show that this condition is satisfied for the important special cases of (i) players’ cost functions being concave in their own continuous strategies with polyhedral strategy sets and (ii) the players’ cost functions only depending on their own strategy and the rivals’ integer strategy components. Finally, we present numerical results on various NEPs and GNEPs. In particular, we show that our tailored B&C method outperforms the direct application of the state-of-the-art bilevel solver MibS on the tested integer and generalized knapsack instances.