The focus of this triple‐blind study was on evaluating the effect of chitosan combined with Dysphania ambrosioides (A) extract on the bone repair process in vivo. In total, 60 male Wistar rats (Rattus norvegicus albinus) weighing between 260 and 270 g were randomly selected for this study and distributed into four groups (n = 15). Group C (chitosan), Group CA5 (chitosan + 5% of D. ambrosioides), Group CA20 (chitosan + 20% of D. ambrosioides), and Group CO (Control‐Blood clot). In each animal, bone defects measuring 2 mm in diameter were performed in both tibias for placement of the substances. After 7, 15, and 30 days, the animals were sedated and sacrificed using the cervical dislocation technique and the tissues were analyzed under optical microscope relative to the following events: inflammatory infiltrate, necrosis, osteoclasts, osteoblasts, fibroblasts, periosteal, and endosteal bone formation. The data were evaluated to verify distribution using the Kolmogorov–Smirnov test, and variance, using the Levene test; as distribution was not normal, data were subjected to the Kruskal–Wallis and Dunn nonparametric tests (p < .05). A significant inflammatory infiltrate was observed in Group CA5 (p = .008) in the time interval of 7 days, and in Group C at 15 (p = .009) and 30 (p = .017) days. Osteoblastic activity was more significant in Group CA20 (p = .027) compared with CA5 in the time interval of 7 days. Group CA20 demonstrated a significantly higher endosteal and periosteal bone formation value in the time interval of 7 (p = .013), 15 (p = .004), and 30 days (p = .008) compared with the other groups. The null hypothesis was refuted, bone regeneration was faster in spheres with an association of chitosan and 20% extract, and complete bone repair occurred clinically at 15 days and histologically at 30 days. The spheres proved to be a promising method for the biostimulation of alveolar bone repair and bone fractures.