The researchers propose a simplified agentic modeling method for simulating large societies of large language model (LLM) agents. Since simulating such societies is expensive, yet the questions asked of such simulations are usually macroscopic: phase behavior, stylized facts, and scaling with the number of agents $N$, not the cognition of any single agent. They turn a statistical-physics observation into a method: replace each LLM agent by a low-parameter model fitted from a few hundred to a few thousand cheap queries, then run the society at any $N$ on a laptop. Whether this works is decided before the simulation runs, chiefly by what each agent perceives. They introduce an [interaction order x memory] taxonomy that maps perception and memory to an effective theory and a predicted $N$-trend of the surrogate error. They validate it on a faithful reimplementation of the LLM macroeconomy EconAgent and seven further named LLM simulations, with agent decisions cloned from genuine LLM elicitations (primarily DeepSeek) for a few dollars; the predicted error trends hold cell by cell, and the two refuted predictions, both on a strongly saturating response and traced to its curvature, are themselves matched quantitatively by the theory with no free parameters. Blogger's Review: This study provides an efficient solution for simulating large language model agent societies by simplifying agent models and perception mechanisms, enabling large-scale simulations to run on a laptop, which is of great significance for language model research and applications.