Dual/tri-GPU multi-agent simulation framework. Orchestrates LLM-powered agents across GPU slots with FLAME population dynamics, Bayesian calibration (ABC-SMC), sensitivity analysis (Morris/Sobol), chaos metrics (Lyapunov, transfer entropy), and graph-based agent memory. Domain-agnostic: swap a YAML scenario file to simulate any multi-agent system.
Fund this project
Gianson
github.com/Giansn/dualmirakl
github.com/Giansn/dualmirakl.git