Phytoremediation is a technique with high viability due to its low cost and minimal environmental impact. However, there are a few results regarding the use of tropical grasses in the phytoremediation of manganese (Mn), and there is no clear definition of which extractor is more suitable for assessing Mn availability in soils. The objective of this study was to evaluate the growth potential and phytoremediation of five tropical grasses in Mn‐contaminated soil, as well as to assess the efficiency of four extractors in Mn extraction. The grass cultivars evaluated included Aruana and Tanzânia of Megathyrsus maximus, Xaraés and Marandu of Urochloa brizantha, and Basilisk of Urochloa decumbens grown in Typic Hapludox (Oxisol) amended with 0, 50, 150, and 450 mg of Mn per kg−1 of soil. The experiment was conducted in a greenhouse, using a completely randomized experimental design in a 5 × 4 factorial scheme, with five replications. The experimental period lasted for 90 days. The dry matter of the shoot and root of the grasses was evaluated. Mn concentration in dry matter and its availability in soil were extracted by Mehlich‐1 and DTPA pH 7.3, while the semitotal concentration was determined by USEPA 3051 and total Mn by USEPA 3052 method. Tropical grasses showed reduced growth with increasing Mn rates in the soil, with the following sequence in the reduction in growth: Aruana > Basilisk > Marandu > Xaraés and lastly the Tanzânia. The Aruana cultivar showed greater potential for phytoextraction due to its higher ability to accumulate Mn in its biomass, followed by the Basilisk cultivar. The four different evaluated extractors were efficient in measuring Mn concentrations in the soil, with higher toxicity critical levels for the USEPA 3051 method due to the semitotal extraction nature and USEPA 3052 due to total Mn extraction. The growth of tropical grasses is reduced with increasing Mn rates applied to the soil. The Aruana of M. maximus grass has greater potential for phytoextraction of Mn‐contaminated soils, followed by Basilisk of Urochloa decumbens, Xaraés and Marandu of U. brizantha, and Tanzânia of M. maximus. The Mehlich‐1, DTPA pH 7.3, USEPA 3051, and 3052 extractors were efficient in extracting and evaluating Mn availability in soil for tropical grasses. The USEPA 3051 and 3052 methods presented higher toxicity critical levels.