Body temperature is an important tool for assessing animal health and welfare. However, there is still limited information on temperature variation in animal models of human pathologies. This study aims to assess whether infrared thermography can be used to follow body temperature variations in a mouse model of sepsis-induced by cecal ligation and puncture (CLP) and compare it with the readout from passive integrated transponder (PIT) tags. We report data retrieved from 27 C57BL/6 wildtype (WT) that underwent severe cecal ligation and puncture or sham surgery, as well as 8 mice subjected to a less severe form of the intervention: 12 Knockout mice on a C57BL/6 background and 12 WT mice. All mice were monitored 4 times per day in the severe model and 3 times per day in the midgrade CLP by PIT tag and infrared thermography for 10 days post-surgery, or until reaching a humane endpoint. Mean body surface temperature (MBST) assessed by infrared thermography and subcutaneous temperature (SCT) were correlated, albeit not strongly. MBST does not appear to be a reliable predictor of non-recovery stages. SCT and percentage of weight loss show some promise to reliably predict non-recovery stages in a severe model of CLP induced sepsis.