This repository contains agents and environments described in the ICML'18 paper "Synthesizing Programs for Images using Reinforced Adversarial Learning". For the time being, we are providing two simulators: one based on libmypaint and one based on Fluid Paint (NOTE: our implementation is written in C++ whereas the original is in javascript). Additionally, we supply a Sonnet module for the unconditional agent as well as pre-trained model snapshots (9 agents from a single population for libmypaint and 1 agent for Fluid Paint) available from TF-Hub.