Towards Agentic Agent-based Models: Feasibility, Performance, and Statistical Model Checking
2026-07-20 • Artificial Intelligence
Artificial IntelligenceMultiagent Systems
AI summaryⓘ
The authors studied how adding large language models (LLMs), which understand and generate text, affects agent-based models (ABMs), especially a model that simulates neighborhood segregation. They created a version where one agent uses an LLM to decide if neighbors are similar or different, while others use simple rules. The tests showed that smaller LLMs sometimes made errors or got stuck, but larger ones worked better. They also used statistical techniques to measure how adding LLMs changes the model's behavior and cost. This helps understand if and how LLMs can be reliably included in simulations of complex systems.
Agent-based models (ABMs)Large language models (LLMs)Schelling segregation modelStatistical model checkingMesa Python libraryMultiVeStASemantic classificationTool callsHybrid agents
Authors
Stefano Blando, Emanuele Guerrazzi, Riccardo Porcedda, Giuseppe Squillace, Max Tschaikowski, Andrea Vandin
Abstract
Agent-based models (ABMs) rely on simple, explicit and reproducible rules for individual decision making, while complex collective behavior emerges from interactions among agents. Recent advances in large language models (LLMs) make it tempting to replace, enrich, or perturb these rules with LLM-based agentic capabilities. However, this raises a methodological question: how does introducing LLM-driven decisions affect the reliability, computational cost, and behavior of ABM simulations? We investigate this for Mesa ABM models, a popular Python library for ABMs, analyzed by statistical model checking. Building on Mesa's integration with the statistical model checker MultiVeStA, we extend the classical Schelling segregation model with a hybrid population: ordinary agents classify neighbors using the standard symbolic rule, while one agent delegates this task to an LLM through tool calls. The LLM-enabled agent receives natural-language descriptions of neighboring agents and invokes tools that increment counters of similar/different neighbors; these counters determine its happiness according to the original Schelling dynamics. This provides a minimal but controlled setting where the semantic, operational, and computational behavior of LLM-based decisions can be studied inside an otherwise standard ABM. We report preliminary experiments with locally served LLMs of different sizes, showing that smaller models may fail simple semantic classification experiments or become operationally unusable during repeated tool-call generation, while larger tested models pass these preliminary checks. We discuss how statistical model checking can estimate classical ABM observables and quantify the impact of introducing agentic LLM components into simulation models.