Toxoplasma gondii is a zoonotic parasite of worldwide distribution. The consumption of infected pork meat has been suggested to be an important source for human infection in the tropical area of Yucatan, Mexico. We performed a cross-sectional study of 12 farms across the state to investigate the seroprevalence of Toxoplasma gondii infection in domestic pigs. In total, 632 samples were obtained from 2 different environmental zones (tropical deciduous low forest and tropical sub-deciduous medium forest) and 2 abattoirs. The modified agglutination test (MAT) was used to assess the seroprevalence of T. gondii in pigs and to evaluate 2 globally used serological tests, the Dye test (DT) and ID Screen® ELISA multi-species, and a commercial ELISA kit (Human Toxo IgG, Human-diagnostics), which is widely used locally in this geographical area. The overall prevalence obtained with the MAT (cut-off ≥1:25) among the 632 pigs was 1.4% (95% CI, 0.6–2.7%). The seroprevalence obtained for the different age groups was 0.6%, 0.7%, 1.8%, and 6.8% among 2–3, 3–4, 4–5, and ≥5-mo-old pigs. This increase in the seroprevalence was statistically significant for the 2 older groups (odds ratio [OR] 3.9–7.1, P < 0.05) in comparison with younger groups. DT at >4 IU dilution had a perfect agreement and 100% of sensitivity and specificity when compared with the MAT. Although ID Screen® had only a fair agreement (κ = 0.389) with the MAT, the McNemar test showed that the results of these tests were comparable (P = 0.29). The Human Toxo ELISA showed no agreement with MAT, ID Screen®, and DT (κ = 0.000–0.023, McNemar P < 0.05). This ELISA was lacking in specificity, accuracy, and precision; hence, we do not recommend its use for T. gondii diagnosis in pig serum.