Objective To investigate the occurrence of malnutrition in patients with gastrointestinal malignancy according to the Global Leadership Initiative on Malnutrition criteria GLIM and compare the consistency of GLIM and patient-generated subjective global assessment PG- SGA in the diagnosis of malnutrition. Method A total of 220 patients with gastrointestinal malignancies admitted to the First Affiliated Hospital of Anhui Medical University were selected from October 2021 to July 2022 for nutritional assessment using the GLIM criteria and PG-SGA. Patients' body composition was measured by bioelectrical impedance technology and Kappa consistency test was used to analyze the consistency of GLIM and PG-SGA evaluation results. Result 112 patients 50. 9% were diagnosed as malnutrition by GLIM including 85 moderate and 27 severe malnutrition. The severity of malnutrition was significantly different in age weight body mass index right upper arm circumference right calf circumference hand grip strength fat mass fat mass percentage fat free mass fat free mass index muscle and skeletal muscle P < 0. 05 . One hundred and 58 patients 71. 8% were diagnosed as malnutrition by PG-SGA including 85 moderate and 73 severe malnutrition. GLIM and PG-SGA had good consistency in the diagnosis of malnutrition Kappa = 0. 548 P < 0. 001 and the area under the ROC curve was 0. 821 95% CI = 0. 745-0. 877 . When distinguishing patients with different degrees of malnutrition the level of consistency was relatively low Kappa = 0. 344 P<0. 001 . Conclusion GLIM criteria and PG-SGA have a good consistency in the diagnosis of malnutrition but a relatively poor consistency in the classification of the severity of malnutrition. However this does not affect the clinical applicability of GLIM criteria.