Background Severe Acute Malnutrition (SAM) has become a major public health challenge in developing countries including Ethiopia, especially among the underprivileged population. Ethiopia is among the developing countries with the highest burden of acute malnutrition among under-five children. Though, plenty of studies were done on the magnitude of acute malnutrition among under-five children in Ethiopia, there is a limited evidence on time to recovery from SAM and its predictors among children aged 6–59 months in Ethiopia, particularly in the study area. Objectives The study was aimed to assess the time to recovery from SAM and its predictors among children aged 6–59 months at Asosa general hospital (AGH), Benishangul Gumuz, Ethiopia. Methods A Five years retrospective follow-up study design was employed among 454 children admitted with SAM in AGH from January 2015 to December 2019. The data were extracted from the patient medical records using checklist. The data were coded and entered into Epi-Data 3.1; then exported to STATA/SE-14 for analysis. Proportional Cox regression was performed to identify predictors of recovery time. A proportional hazard assumption was checked. Variables with AHR at 95% CI and P-value less than 0.05 in the multivariable Cox proportional regression was considered as significant predictors of recovery time. Findings Among the 454 included records of children with SAM, 65.4% (95%CI: 50.1, 69.2) of them were recovered at the end of the follow-up with a median recovery time of 15 IQR(11–18)days. The incidence rate of recovery was 5.28 per 100 child days’ observations. Being HIV Negative (AHR = 2.19: 95% CI 1.28, 3.73), Marasmic (AHR = 1.69: 95% CI 1.18, 2.42), and marasmic-kwashiorkor child (AHR = 1.60: 95% CI (1.09, 2.37) independently predicted recovery time. Conclusions Though the time to recovery from severe acute malnutrition was in the acceptable range, the proportion of recovery was found to be low in the study area compared to sphere standard. The prognosis of children with severe acute malnutrition was determined by the HIV status of the child and the type of malnutrition experienced. Further strengthening of malnutrition therapeutic centers and routine checkup of the nutritional status of HIV positive children should be emphasized to reduce child mortality and morbidity from under-nutrition.