Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a Laptop

Researchers introduced a method to simulate large societies of large language model (LLM) agents on a laptop by replacing each agent with a low-parameter model fitted from a few hundred to a few thousand cheap queries, validated on a reimplementation of the LLM macroeconomy EconAgent and seven other named LLM simulations. The method, based on an interaction order x memory taxonomy, predicts error trends that held cell by cell, with two refuted predictions on saturating responses matched quantitatively by the theory with no free parameters.

arXiv:2608.11215v1 Announce Type: new Abstract: Simulating societies of many large language model LLM agents is expensive, yet the questions asked of such simulations are usually macroscopic: phase behaviour, stylised facts, and scaling with the number of agents $N$, not the cognition of any single agent. We 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. We 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. We 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.