Background: Molecular classification of endometrial cancer has been applied in patient stratification and treatment strategy. Although the Cancer Genome Atlas project and other genomic studies have established comprehensive molecular classification, its genomic characteristics in Chinese endometrial cancer remain unclear. Methods: Tumor tissue and matched white blood cells from 316 patients with endometrial cancer underwent next-generation sequencing (NGS), which targeted the whole coding region of 733 or 381 genes. Endometrial cancer was classified into four subtypes according to genomic characteristics: polymerase (POLE) ultramutated, microsatellite instability (MSI), TP53 wild type (TP53wt), and TP53 mutant (TP53mut). Tumor mutation burden (TMB) was calculated based on the number of somatic single nucleotide variants and indels in examined coding regions. Deleterious or suspected deleterious genes were included for analysis. Genomic instability was also evaluated with Homologous Recombination Deficiency (HRD) score by over 10, 000 single-nucleotide polymorphisms distributed across human genome. HRD algorithm combined loss of heterozygosity score (LOH), telomeric allelic imbalance score (TAI) and large-scale state transition score (LST). Results: The 316 endometrial cancer patients were divided into four subtypes: POLE ultramutated (16, 5.06%), MSI tumors (70, 22.2%), TP53wt (131, 41.5%), and TP53mut (99, 31.3%). Median TMB was 26.5 mut/Mb in overall cohort and POLE ultramutated subtype possessed the highest TMB (181 mut/Mb), followed by MSI (54.1 mut/Mb), TP53mut (7.11 mut/Mb) and TP53wt (6.61 mut/Mb) subtypes. POLE, PTEN, LRP1B, BRCA2 and ATRX were the top frequently mutated genes in POLE ultramutated subtype, which mutation rate are more that 80%. The most frequent mutations were detected in ARID1A, PTEN, PIK3CA in MSI subtype, TP53, PIK3CA, PTEN in TP53mut subtype, and PTEN, PIK3CA, ARID1A in TP53wt subtype. HRD score was significantly higher in TP53mut subtype, followed by TP53wt, MSI and POLE ultramutated subtype (median HRD score, 21.7, 6.15, 3.65 and 0; P < 0.001). Conclusions: Our study revealed the heterogeneous genomic characteristics of four molecular subtypes in Chinese endometrial cancer. POLE ultramutated and MSI subtypes display higher TMB. TP53mut subtype showed genome features of HRD and would likely benefit from HRD-targeted therapy.