Service crops (or cover crops) play an important role in simplified agricultural systems. Service crops reduce agricultural external inputs and increase ecosystem services but their ability to mitigate nitrous oxide (N2O) emissions is still uncertain. The main objective of this study was to evaluate N2O emissions in soybean–soybean (Glycine max [L.] Merr) rotations that included different service crops. Treatments included continuous soybean with winter fallow and soybean with three service crops: oat (Avena sativa L.), vetch (Vicia villosa Roth.), and a mixture of oat and vetch in a randomized complete block design. Service crops were sown 2 months after soybean harvest and were terminated 2 months before soybean planting. Nitrous oxide emissions were determined during the fourth year of the field experiment. We found that service crops did not significantly affect overall mean N2O emission rates, with mean emission rates from the fallow, oat, vetch, and oat‐vetch treatments of 1.82 ± 0.35, 1.95 ± 0.34, 2.71 ± 0.43, and 2.42 ± 0.42 kg N2O‐N ha−1 per year, respectively. Service crops with low C/N ratios (vetch and oat‐vetch mixtures) significantly increased N2O emissions in spring, after their termination. Overall, soil inorganic N content (NO3− or NH4+) was the main driver that explained the N2O emissions from different treatments, whereas water‐filled pore space controlled the temporal variability of emissions. Our results suggest that service crops with a very short growing season may increase soil N availability for cash crops, but do not reduce N2O emissions due to long periods of high N availability without crops.