Rice is a cereal that presents a great ability to adapt to different soil and climate conditions. However, as it is a tropical crop with C3 metabolism, it performs better in warm temperatures with high solar radiation. Tolerance to stress caused by low temperatures is a highly complex process that involves various metabolic pathways and cellular compartments, resulting in general or specific effects on plant growth and development. In order to observe the true effect of a particular stress on genetic expression, reference genes need to be chosen for real-time PCRs, the expression levels of which should remain stable independent of the situation imposed. In this paper, the expression stability was evaluated of the actin 11 (ACT11), ubiquitin-conjugating enzyme 2 (UBC-E2), glyceraldehyde 3-phosphate dehydrogenase (GAPDH), beta tubilin (β-Tubulin), eukaryotic initiation factor 4α (eIF-4-α), eukaryotic initiation factor 1α (eIF-1-α), ubiquitin 10 (UBQ10), ubiquitin 5 (UBQ5), aquaporin (TIP41), and cyclophilin genes, in two rice genotypes cultivated in low temperature (13°C) conditions in vegetative stage (V4). The analysis material (leaves) was collected after 0, 6, 24, 48, and 72 h of exposure to the stress. In this study, the geNorm, BestKeeper, ΔCt, NormFinder, and RefFinder methods were used to evaluate the expression stability of the candidate reference genes. The results revealed that the most indicated genes for all the analysis methods were UBQ10 and UBQ5 for BRS Bojuru and BRS Pampa, respectively. On the other hand, the eIF-1-α gene presents the least expression stability and is not indicated for studies of rice plants subjected to low temperatures. The validation with the antioxidant system genes SODCc1-Cu/Zn, CATC, APX2, and GR2 confirmed the importance of using previously tested normalizing genes for adequate real-time PCR results.