Animation is pervasive on the Web, the CSS animation property alone appears on over three-quarters of webpages. Yet existing webpage code generation benchmarks target static screenshots, user-triggered interactions, or isolated motion graphics, leaving the ability of multimodal large language models (MLLMs) and coding agents to reconstruct complete animated webpages largely unexplored. To fill this gap, we introduce AnimWeb-Bench, the first benchmark for animated webpage code generation from videos. It comprises 124 real-world webpages spanning 7 application domains and 33 animation types, with 91.9% of pages combining multiple types, each paired with a reference video and the static assets of the original webpage. We further propose a human-aligned, two-dimensional evaluation protocol: Static Appearance Fidelity measures structure, text, color, and visual similarity on aligned keyframes, while Animation Fidelity tracks UI components across frames and compares their position, size, velocity, and appearance with those in the reference video. Evaluating 14 MLLMs and three coding agents reveals: (1) reproducing webpage animations is substantially more difficult than reproducing static appearance; (2) explicit frame inspection and render feedback improve the reconstruction quality of coding agents; and (3) rapid, continuous changes are harder to reproduce than discrete state transitions. Our code and dataset are available at https://starry1001.github.io/AnimWeb-Bench-website/ .
Figures & tables
Figure 1. Overview of the AnimWeb-Bench task and evaluation. Given a reference video and the static assets of a real webpage (top), an MLLM or coding agent generates a single HTML file, which is rendered and recorded in a browser (bottom). Static Appearance Fidelity compares aligned keyframes. Animation Fidelity compares the trajectories of matched UI components.
Benchmark
Input
Data Source
Samples
Target Domain
Pix2Code ( Beltramelli, 2018 )
Image
Synthesized
1742
Static Layouts
Web2Code ( Yun et al., 2024 )
Image
Synthesized
60K
Static Layouts
WebSight ( Laurençon et al., 2024 )
Image
Synthesized
2M
Static Layouts
Design2Code ( Si et al., 2025 )
Image
C4
484
Static Layouts
WebCode2M ( Gui et al., 2025a )
Image
Common Crawl
2.56M
Static Layouts
CC-HARD ( Gui et al., 2025b )
Image
Common Crawl
128
Static Layouts
Table 1. Overview of existing UI-to-code benchmarks.
Figure 2. Construction pipeline of AnimWeb-Bench.
Type
Number
Type
Number
Type
Number
Type
Number
Translation
64
Fade Out
12
Blur Transition
4
Font Weight Animation
2
Background Transition
50
Rotation
11
Icon State Animation
4
Horizontal Image Slide
2
Appearance
49
Numeric Counter Animation
10
Shake
4
Vertical Image Slide
2
Masking
32
Scale Down
8
3D Depth Motion
3
Logo Morphing
1
Fade In
30
Text Substitution
7
Grid Tile Transition
3
Radial Expansion Animation
1
Text Animation
24
Image Stack Animation
5
Icon State Transition
3
Text Fragmentation
1
Table 2. Distribution of animation types in AnimWeb-Bench.
Figure 3. Domain distribution of AnimWeb-Bench.
Figure 4. Keyframe extraction on an example from AnimWeb-Bench. The last two detected keyframes have nearly identical content and are therefore merged into k4 .
Table 3. Performance comparison. All metrics are higher-is-better. Models marked with † receive the native video, and the others receive frames sampled at 2 fps. The best and second-best scores are marked in bold and underlined, respectively.
While recent vision-language models (VLMs) have achieved significant improvements on static visual-to-code tasks such as generating code for webpages, charts, or SVGs, it remains unclear whether they can recover temporal dynamics when motion is present. To this end, we introduce Animation2Code, a benchmark for evaluating temporal visual reasoning via reconstructing executable web animation code from videos. Animation2Code consists of 1,069 web animation videos with diverse visual appearances and motion patterns, paired with corresponding HTML/CSS/JavaScript implementations. We propose two human-aligned metrics, appearance similarity and temporal similarity, which allow us to disentangle visual fidelity from temporal alignment when comparing rendered animations against ground-truth samples. Benchmarking state-of-the-art VLMs on this dataset shows that current VLMs struggle to maintain temporal consistency in reconstruction, even when achieving high appearance similarity, including under finetuning and iterative refinement settings. Code and data are available at https://anya-ji.github.io/animation2code-website .
Anya Ji, Abhijith Varma Mudunuri, David M. Chan +1
Existing web-generation benchmarks rely on text prompts or static screenshots as input. However, videos naturally convey richer signals such as interaction flow, transition timing, and motion continuity, which are essential for faithful webpage recreation. Despite this potential, video-conditioned webpage generation remains largely unexplored, with no dedicated benchmark for this task. To fill this gap, we introduce WebVR, a benchmark that evaluates whether MLLMs can faithfully recreate webpages from demonstration videos. WebVR contains 175 webpages across diverse categories, all constructed through a controlled synthesis pipeline rather than web crawling, ensuring varied and realistic demonstrations without overlap with existing online pages. We also design a fine-grained, human-aligned visual rubric that evaluates the generated webpages across multiple dimensions. Experiments on 19 models reveal substantial gaps in recreating fine-grained style and motion quality, while the rubric-based automatic evaluation achieves 96% agreement with human preferences. We release the dataset, evaluation toolkit, and baseline results to support future research on video-to-webpage generation.
Recent advancements in multimodal large language models (MLLMs) have achieved remarkable progress in multimodal reasoning and code generation, catalyzing a new paradigm for front-end development. In particular, these models can directly transform visual designs into executable code, significantly improving the efficiency and adaptability of web development. Modern web applications are dynamic and interactive, featuring frequent user-page interactions. However, existing benchmarks largely evaluate the code generation of static webpages, ignoring the complex interactive behaviors in real-world applications. Besides, their evaluation criteria remain confined to visual fidelity and code structure, overlooking the interaction consistency between the generated and the reference webpages. To address these limitations, we introduce WebIGBench, the first benchmark designed to evaluate code generation for interactive webpages with complex interactions. By combining manually designed interaction paths with UI automation, we collected 103 complex webpages from real-world websites. This benchmark covers 5 popular interactive action types (e.g., click, input) involving 871 distinct interactive actions. Moreover, we propose a novel evaluation pipeline to address the gap in automated assessment of interactive actions. Extensive experiments on several representative MLLMs reveal the performance boundaries of current models in interactive webpage code generation using WebIGBench. The proposed benchmark is available at https://github.com/anoa12159-hue/WebIGBench_eval.
Fan Wu, Lishuai Dong, Cuiyun Gao +4
Computer Science and Technolog, Harbin Institute of Technology. · Institute of Information Engineering, Chinese Academy of Sciences.