The objective of this work was to evaluate if different rate of a nicosulfuron and atrazine mixture applied post-emergence to effectively control weeds in sweet sorghum without interfering in the initial growth and gas exchanges of the crop. The genotypes BRS 506, BRS 511, and SF 15 were subjected to weed management with atrazine (2500 g a.i. ha−1) mixed with six different doses of nicosulfuron (10, 20, 30, 40, 50, 60 g a.i. ha−1). The herbicide mixtures were applied 12 days after sowing. At 7, 14, 21, and 28 days after application (DAA), phytotoxicity evaluations were performed in sorghum genotypes and weed control efficiency was evaluated. Initial growth (plant height, stalk diameter, leaf area, and number of leaves), gas exchanges, and relative chlorophyll index (a and b) were determined 7 DAA. The production of total fresh biomass, total soluble solids and ethanol yield were carried out at the end of the culture cycle. The mixture with doses higher than 40 g a.i. ha−1 of nicosulfuron with atrazine caused greater phytotoxicity in sweet sorghum and better weed control. Gradual increase in nicosulfuron reduced photosynthesis, transpiration, and initial growth of sorghum crop. The weed control did not interfere with production of fresh biomass, total soluble solids content and ethanol production, but showed different response in tested genotypes (BRS 511 > BRS 506 > SF 15). Thus, within the dose range evaluated, the mixture of 30 g a.i. ha−1 of nicosulfuron with atrazine was found effective in controlling weeds in sweet sorghum plantations without causing significant damage to the crop.