This chapter uses a swarm-intelligence-based approach to solve transmit beamforming problems in wireless communications. An extensive review has been carried out about transmit beamforming designs and the generalized Firefly Algorithm (FA) introduced by Le and Yang in 2023 as an optimization framework for solving transmit beamforming problems. Such design problems are subject to multiple, nonlinear multivariate constraints with matrix inputs. The proposed approaches are described in detail and two transmit beamforming problems are then used to illustrate the effectiveness of the generalized Firefly Algorithm to find optimal solutions. Semidefinite relaxation (SDR) technique is also utilized to solve the same problems for comparison with the results obtained by the FA approaches. Comparison and analyses reveal the fact that when the number of antennas is large, the proposed FA approaches have less computational complexities, compared with those of their SDR counterparts. Numerical results also indicate that the FA approach can obtain the same global optimal design solution as the SDR does for a hidden-convex problem in transmit beamforming. Furthermore, the FA approach can outperform its SDR counterpart in terms of attaining better optimal solutions for non-convex, multivariate optimization problems in wireless communications.