Cardiovascular magnetic resonance imaging (MRI) has gained widespread acceptance for the assessment of cardiovascular disease. Cardiac MRI requires fast data acquisition schemes because of constraints imposed by physiological motion of cardiac structures and blood flow, which dictate the suitable window of data acquisition. The ongoing improvement of MRI hardware and the development of tailored imaging techniques have been the cornerstones for rapid progress in cardiac MRI. Cardiac MRI at 3.0 T holds the promise to overcome some of the signal-to-noise (SNR) limitations, especially for techniques with borderline SNR at 1.5 T (eg, myocardial perfusion, assessment of viability, or imaging of coronary arteries). The improved SNR at 3.0 T can be used to increase the spatial resolution and/or reduce imaging time. It was shown that all applications of cardiac imaging at 1.5 T seem feasible also at 3.0 T and predominantly provide similar or improved image quality. Although specific absorption rate limitations and susceptibility effects remain a primary concern, the combination of high-field strength examinations with parallel imaging has increased the performance of techniques such as steady-state free-precession at 3.0 T. Therefore, the signal-to-noise and the contrast-to-noise ratios advantages at 3.0 T and the resulting potential benefit for an improved diagnostic value will constantly fuel further developments in this area and pave the way for novel, promising imaging techniques.