Species with different genetic backgrounds exhibit distinct metabolic traits. Nine beef cattle were selected for the experiment to study changes in serum metabolic phenotypes, rumen microbiota diversity, and composition in beef cattle from different genetic backgrounds. Three groups were Chinese Simmental (S group), Simmental×Chinese Holstein (SH group), and Simmental × Mongolian (SM group) cattle. We used ELISA to detect serum biochemical indicators. The Short-chain fatty acids (SCFAs) in the rumen were examined, and a significant difference was observed in the acetic acid content of the three experimental groups (p < 0.01). The propionic acid content in the rumen of the S group was significantly higher than that of the SH and SM groups (p < 0.05). The A/P ratios of both the S and SM groups were significantly higher than that of the SH group (p < 0.05). We analyzed rumen microbiota composition and diversity in each group of cattle using 16 S rRNA sequencing and found that their composition was generally similar in the three groups of crossbred fattening cattle; however, the f_Bacteroidales_RF16_group and g_norank_f_Bacteroidales_RF16_group were significantly enriched in the SH group, whereas Treponema and Spirochaetia were significantly enriched in the SM group. Spirochaetia was significantly enriched in the SM group. Differences in rumen bacterial enrichment indicated that starch, protein, and cellulolytic abilities differed among the S, SH, and SM groups. The results of Spearman correlation analysis confirmed the correlation between rumen genera and serum biochemical indices. Overall, differences in rumen microflora play an important role in influencing the serum metabolic phenotype.