Background: Non-optimum ambient temperature has not been widely perceived as an important environmental risk factor for asthma, and the association between ambient temperature and pulmonary function is rarely explored. Our study aimed to investigate the associations between non-optimum ambient temperature and pulmonary function among asthmatic adult patients. Methods: We performed a longitudinal study among 4,992 eligible adult asthmatic patients in 25 cities of China from 2017 to 2020. The patients were required to complete pulmonary function test every day in the morning and evening. Linear mixed-effects models and distributed lag non-linear models were used to evaluate the associations between ambient temperature and pulmonary function. Results: We evaluated 298,396 records of pulmonary function tests. We found inversely J-shaped exposure-response relationship curves for ambient temperature and pulmonary function. The effects of extreme low temperature occurred at lag 0 h and vanished at lag 72 h (almost 3 days). Compared with referent temperature (29.5 degrees C), extreme low temperature (9.4 degrees C) was associated with decreases of 60.4 mL in FEV1, 299.7 mL/s in PEF, and 101.5 mL in FVC. Extreme high temperature (34.2 degrees C) was associated with decreases of 26.0 mL in FEV1, 35.8 mL/s in PEF, and 23.4 mL in FVC. Patients of male, overweight, and elder ages were vulnerable populations, and cold effects were more prominent in the south and in areas without central heating. Conclusions: Both extreme low and high ambient temperatures were associated with decreased pulmonary function in adult asthmatic patients. The effect could last for almost 3 days and low temperature was more harmful.