GEBench: Benchmarking Image Generation Models as GUI Environments

TL;DR

GEBench introduces a comprehensive benchmark and a novel metric for evaluating dynamic, multi-step GUI image generation models, emphasizing temporal coherence and interaction logic, revealing current models' limitations in long-term consistency.

Contribution

This work presents GEBench, a new benchmark with a five-dimensional evaluation metric for assessing GUI image generation models' temporal and interaction fidelity, addressing a significant evaluation gap.

Findings

Models perform well on single-step transitions.

Models struggle with temporal coherence in multi-step sequences.

Icon interpretation and localization are key bottlenecks.

Abstract

Recent advancements in image generation models have enabled the prediction of future Graphical User Interface (GUI) states based on user instructions. However, existing benchmarks primarily focus on general domain visual fidelity, leaving the evaluation of state transitions and temporal coherence in GUI-specific contexts underexplored. To address this gap, we introduce GEBench, a comprehensive benchmark for evaluating dynamic interaction and temporal coherence in GUI generation. GEBench comprises 700 carefully curated samples spanning five task categories, covering both single-step interactions and multi-step trajectories across real-world and fictional scenarios, as well as grounding point localization. To support systematic evaluation, we propose GE-Score, a novel five-dimensional metric that assesses Goal Achievement, Interaction Logic, Content Consistency, UI Plausibility, and Visual Quality. Extensive evaluations on current models indicate that while they perform well on single-step transitions, they struggle significantly with maintaining temporal coherence and spatial grounding over longer interaction sequences. Our findings identify icon interpretation, text rendering, and localization precision as critical bottlenecks. This work provides a foundation for systematic assessment and suggests promising directions for future research toward building high-fidelity generative GUI environments. The code is available at: https://github.com/stepfun-ai/GEBench.