Objective To explore the risk factors for early-onset sepsis (EOS) in extremely low birth weight (ELBW) premature infants, and construct a nomogram model for EOS in ELBW premature infants. Methods A total of 200 ELBW premature infants who admitted to a hospital from January 2021 to December 2023 were selected as the training set, and 86 ELBW premature infants who admitted to the hospital from January to December 2024 were selected as the validation set. The ELBW premature infants in the training set were divided into an EOS group and a non-EOS group based on the occurrence of EOS. The risk factors for EOS in ELBW premature infants were screened out by logistic regression, and a nomogram model for EOS in ELBW premature infants was constructed using R-based software. Results Maternal gestational age >35 years, prenatal fever, premature rupture of membranes, peripherally inserted central venous catheter (PICC) insertion, mechanical ventilation, amniotic fluid contamination, gestational age ≤32 weeks, and neonatal fever were independent risk factors for EOS in ELBW premature infants (all P<0.05). The area under the receiver operating characteristic (ROC) curve for the training set and validation set were 0.797 (95%CI: 0.755-0.859) and 0.769 (95%CI: 0.661-0.877), respectively. The calibration curve showed that the model had good consistency, and the decision curve showed that the model had high clinical application value. Conclusion The dynamic nomogram model for predicting EOS in ELBW premature infants has good accuracy and clinical practicality.