The objective of this work was to evaluate the effect of cover crops on weed suppression in no-tillage soybean production systems in Rondonopolis, Mato Grosso. The experiment was carried out in an experimental area and consisted of the evaluation of nine cover treatments and soil management in a randomized complete block design. The treatments were: NT fallow, CT fallow, Crotalaria spectabilis, Crotalaria breviflora, maize + Crotalaria spectabilis, Pennisetum glaucum, Urochloa ruziziensis, Cajanus cajan, sunflower + Urochloa ruziziensis, Stylosanthes, Vigna unguiculata, Urochloa brizantha, maize + Urochloa ruziziensis. The evaluations were carried out before the desiccation for soybean sowing in the crops (10/23/2014) and (09/21/2015), before the post-emergence of soybean (09/12/2014) and (12/11/2015) and in the second season (12/06/2015). The useful area was 5 x 5 m and all weeds were counted and identified, but only the four species with the largest population were collected. The weeds evaluated were: Digitaria horizontalis, Digitaria insularis, Porophyllum ruderale and Tridax procumbens. Fallow treatments presented higher weed populations in relation to the others, in all periods of evaluation. Digitaria horizontalis presented the highest phytomass production in most seasons. The production systems with Urochloa ruziziensis, Pennisetum glaucum, Crotalaria spectabilis and intercropped with maize + Urochloa ruziziensis, sunflower + Urochloa ruziziensis and maize + Crotalaria spectabilis were the best alternatives for integrated weed management, reducing the incidence and increasing control of the main species that were detected during the conduction of the experiment.