NeuralOS: Towards Simulating Operating Systems via Neural Generative Models

ICLR 2026 Conference Submission12976 Authors

Keywords: Generative Models, Neural Simulation, Diffusion Models, Graphical User Interfaces

TL;DR: We introduce a neural generative model that simulates operating system interfaces by directly generating screen images from user inputs.

Abstract: We introduce NeuralOS, a neural framework that simulates graphical user interfaces (GUIs) of operating systems by directly predicting screen frames in response to user inputs such as mouse movements, clicks, and keyboard events. NeuralOS combines a recurrent neural network (RNN), which tracks the computer state, with a diffusion-based neural renderer that generates screen images. The model is trained on a dataset of Ubuntu XFCE recordings, which include both randomly generated interactions and realistic interactions produced by AI agents. Experiments show that NeuralOS successfully renders realistic GUI sequences, accurately captures mouse interactions, and reliably predicts state transitions like application launches. Beyond reproducing existing systems, NeuralOS shows that synthesized training data can teach the model to simulate applications that were never installed, as illustrated by a Doom application, and suggests a path toward learning user interfaces purely from synthetic demonstrations.

Primary Area: generative models

Submission Number: 12976