This study aims to construct a novel risk score model based on glycolysis-related genes in osteosarcoma and to build and validate a prognostic model for predicting overall survival of patients with osteosarcoma. The transcriptome data and corresponding clinical data of patients with osteosarcoma were obtained from The Cancer Genome Atlas (TCGA) as the training set, and from Gene Expression Omnibus (GEO) database as the validation set. Univariate Cox regression analysis was used to screen the prognostic glycolysis-related genes. The risk coefficient of each glycolysis-related gene was calculated using LASSO regression analysis. Using the median risk score as the cut-off point, patients were divided into high-risk and low-risk groups. Kaplan–Meier survival analysis was used to determine whether there was a significant difference in the overall survival between the two groups. The nomogram was constructed according to the results of multivariate Cox regression. The C-index was calculated, the calibration chart, clinical decision curve and receiver operating characteristic curve were drawn to evaluate the predictive performance of the nomogram. We performed Gene Ontology and Kyoto encyclopedia of genes and genomics enrichment analysis to explore the potential mechanism of prognostic-related glycolysis genes in osteosarcoma. A total of 88 and 53 cases were obtained from the TCGA and GEO database, respectively. A total of 10 key glycolytic genes related to prognosis were screened out. The Kaplan–Meier survival curve revealed that the overall survival of the high-risk group was significantly shorter than that of the low-risk group. The C indices of the training set and the verification set were 0.882 and 0.828, respectively. Our findings will provide further understanding of clinical prognostic outcomes of osteosarcoma patients.