Koneru, P. et al. 2026. When Does Disaggregated, Cache-Aware LLM Serving Pay Off? An Analytical Crossover Model Validated on Real Compute, Network, and Coordination Measurements. International Journal of Artificial Intelligence and Machine Learning. 6, 8s (Aug. 2026), 1275–1286. DOI:https://doi.org/10.51483/IJAIML.2970.