Lack of understanding the effects of single- and multiple-weed interference on soybean yield has led to inadequate weed management in Primorsky Krai, resulting in much lower average yield than neighboring regions. A 2 yr field experiment was conducted in a soybean field located in Bogatyrka (43.82°N, 131.6°E), Primorsky Krai, Russia, in 2013 and 2014 to investigate the effects of single and multiple interference caused by naturally established weeds on soybean yield and to model these effects. Aboveground dry weight was negatively affected the most by weed interference, followed by number of pods and seeds. Soybean yield under single-weed interference was best demonstrated by a rectangular hyperbolic model, showing that common ragweed and barnyardgrass were the most competitive weed species, followed by annual sowthistle, American sloughgrass, and common lambsquarters. In the case of multiple-weed interference, soybean yield loss was accurately described by a multivariate rectangular hyperbolic model, with total density equivalent as the independent variable. Parameter estimates indicated that weed-free soybean yields were similar in 2013 and 2014, i.e., estimated as 1.72 t and 1.75 t ha−1, respectively, and competitiveness of each weed species was not significantly different between the two years. Economic thresholds for single-weed interference were 0.74, 0.66, 1.15, 1.23, and 1.45 plants m−2 for common ragweed, barnyardgrass, annual sowthistle, American sloughgrass, and common lambsquarters, respectively. The economic threshold for multiple-weed interference was 0.70 density equivalent m−2. These results, including the model, thus can be applied to a decision support system for weed management in soybean cultivation under single and multiple-weed interference in Primorsky Krai and its neighboring regions of Russia.