GUI Generation Benchmark for Dynamic Interaction and Temporal Coherence

GEBench is a benchmark dataset from stepfun-ai for evaluating dynamic interaction and temporal coherence in Graphical User Interface generation. It addresses a gap in existing benchmarks by focusing on state transitions and temporal coherence within GUI-specific contexts. The dataset was last updated on February 25, 2026.

Use Cases

• Evaluate GUI state transition predictions based on user instructions for temporal coherence
• Benchmark image generation models on GUI-specific visual fidelity and interaction dynamics
• Assess model performance on the task categories of text-to-image, image-to-image, and image-text-to-image generation

Strengths

• Focuses on the underexplored evaluation of state transitions and temporal coherence in GUI contexts
• Covers multiple relevant task categories including text-to-image and image-to-image generation
• Dataset is recent, with a last update recorded in February 2026

Limitations

• Specific data scale, such as row count or number of GUI states, is unknown
• Column structure and sample data are unavailable, limiting initial inspection
• Geographic or application scope may be biased towards the 'us' region tag

Provenance

Source

stepfun-ai on Hugging Face

Collection Method

null

Time Range

null

Freshness

Last updated 2026-02-25.

Geography

Region tag indicates 'us'.