Koneru P, Pola SK, Ravi P. When Does Disaggregated, Cache-Aware LLM Serving Pay Off? An Analytical Crossover Model Validated on Real Compute, Network, and Coordination Measurements. IJAIML [Internet]. 2026 Aug. 1 [cited 2026 Oct. 27];6(8s):1275-86. Available from: https://svedbergopen.com/index.php/ijaiml/article/view/2970