Herbicide mixtures are a modern weed management practice as they reduce herbicide application. This study aimes to evaluate the effect of metribuzin, halosulfuron and flumioxazin applied individually and as mixtures (metribuzin:halosulfuron and metribuzin:flumioxazin) on Chenopodium album, Amaranthus retroflexus, and potatoes on biomass and chlorophyll-a fluorescence in 21 experiments. The individual herbicide experimental design was a randomized completely block design with seven doses and three replications. The factorial experiments were performed on the basis of randomized completely block design in three replications for binary mixture exper-iments. Flumioxazin was very potent in controlling C. album, A. retroflexus, and injured potatoes with a 50% effective dose (ED₅₀ ) of 1.21, 0.54, and 12.23 g ai·ha⁻¹, respectively. Both mixtures of metribuzin:halosulfuron and metribuzin:flumioxazin generally showed an antagonistic effect on both weeds and potato in 12 independent experiments. Metribuzin, halosulfuron, and flumioxazin significantly decreased photosystem II activity by decreasing the maximum quantum efficiency (F_v /F_m ). The metribuzin:halosulfuron mixtures almost followed the Additive Dose Model for F_v /F_m, whilst there was an antagonistic effect for the metribuzin:flumioxazin that was closely related to biomass. The results indicated that mixtures were generally antagonistic, and the endpoint choice is pivotal when assessing the joint action of mixtures.