The objective of this three-year study was to investigate the usefulness of flame weeding, spraying with glufosinate-ammonium and nighttime soil tillage for weed suppression in summer savory. The experiment was arranged in a split-plot design with three experimental factors and four replications. The soil was cultivated with a rotary tiller a day before savory sowing or an hour after sunset on the night preceding sowing, while flaming (90 kg propane∙ha–1) and glufosinate-ammonium (600 g∙ha–1) were applied, depending on the year, 12–22 days after sowing, i.e. after the emergence of weeds and the first savory seedlings. Flaming and glufosinate-ammonium killed all weeds growing during the treatment but they did not prevent new weeds from emerging on the following days. Three weeks later, the number of weeds growing on plots weeded with the flame method and sprayed with glufosinate-ammonium was significantly lower by about 63 and 69% in comparison to control, respectively, and it was independent of the time of soil tillage. The studied methods had no effect on weed infestation assessed 3–4 weeks after the first weeding, or on the emergence, plant height and yield of fresh savory herb. The content of oil in dry herb varied from about 1.9% to 2.4% depending on the year while it did not depend on the weeding method. Oil produced by control plants contained 35 compounds making up 99.85% of the total, with the predominant share of carvacrol (45.2–46.8%) and γ-terpinene (34.6–39.9%). Much smaller was the share of α-terpinene (4.8%), p-cymene (2.7–4.0%), myrcene (1.5–2.0%) and α-thujene (1.4–1.8%). The average share of the remaining components did not exceed 1%.