BACKGROUND AND STUDY AIMS Adequate training is required to achieve successful endoscopic ultrasound (EUS)-guided fine-needle aspiration (EUS-FNA). Of the variety of training models currently available, none offers verisimilitude to the tactile feel of puncturing a human lymph node. The aim of the current study was to evaluate a new porcine lymph node model for EUS-FNA training and to evaluate its impact on trainees' performance in patients compared with the literature of other models available. METHODS Two trainees each performed EUS-FNA of 96 lymph nodes in 18 animals with induced lymphadenopathy (mean 1.6 cm [range 0.9-3.5 cm]). Accuracy, speed, adequacy of sampling, and trainees' performance pre- and post-training were measured. Using a questionnaire, data were gathered regarding the effect of training and comfort level in patients. Results were compared with those in the literature. RESULTS Trainees progressed from hands-on assistance to occasional verbal guidance toward the end of animal training. There was good correlation between puncture time and number of EUS-FNA procedures performed in all but the subcarinal location (r = - 0.17). Comparison of trainee performance in patients before and after training showed a reduction in puncture time (P = 0.0014). Questionnaire analysis revealed increased confidence in echoendoscope- and needle-handling. Comparison with other published models supports these results. CONCLUSION Results from the literature and the current study showed that animal training improves trainee performance, confidence, and procedural comfort when returning to patient examinations. The new model produces a realistic response that is similar to EUS-FNA in patients; this experience provides a benefit to endoscopists in terms of improved performance in patients and could be considered for use in accreditation. Due to the small numbers of trainees, larger experiences are needed to confirm training efficacy.