LongWebBench: leaderboard
Recent vision-language models (VLMs) have shown promising progress in generating webpages from visual inputs, yet existing evaluations mainly focus on short, single-screen, and largely static webpages. We introduce LongWebBench, a benchmark for evaluating long-horizon webpage generation from both structural and functional perspectives. LongWebBench contains 490 real-world long webpages for structu
Metric: Overall (self-reported). Source: benchmarklist.com. Saturation forecast: Around April 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 7.22 | #66 |
| 2 | Claude Sonnet 4.5 (Thinking) | 6.74 | #142 (Claude Sonnet 4.5) |
| 3 | Gemini 3 Flash (Preview) | 6.53 | #82 |
| 4 | GPT-5.2 | 6.5 | #110 |
| 5 | Claude Opus 4.5 (20251101) | 6.43 | #74 |
| 6 | GLM-4.6V | 5.96 | #316 |
| 7 | Qwen 3 VL 235B A22B Instruct | 5.85 | #273 |
| 8 | Seed-1.6 | 4.77 | #265 |
| 9 | Qwen 3 VL 8B Instruct | 3.95 | #420 |
| 10 | InternVL3-78B | 3.62 | #352 |
| 11 | GPT-4o | 3.53 | #343 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=longwebbench · How It Works · Data refreshed daily, snapshot 2026-10-20.