LEE JI-SU. GPU Memory Usage Prediction for Generative AI Serving Pipelines with Queue, Latency, and Utilization Signals. Journal of Advanced Computing Systems , [S. l.], v. 6, n. 7, p. 1–15, 2026. DOI: 10.69987/JACS.2026.60701. Disponível em: https://scipublication.com/index.php/JACS/article/view/422.. Acesso em: 5 jul. 2026.