WebCoderBench - Visual Experience: leaderboard
The Visual Experience track of WebCoderBench — visual quality of the page rendered from LLM-generated front-end code; score (0-100).
Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Codex (High) | 93.83 | #127 (GPT-5 Codex) |
| 2 | GPT-5 (High) | 90.62 | #98 (GPT-5) |
| 3 | GLM-4.5 | 87.42 | #275 |
| 4 | DeepSeek V3.1 Terminus | 86.4 | #220 |
| 5 | Gemini 2.5 Pro | 78.08 | #152 |
| 6 | Qwen 3 235B A22B Instruct | 73.35 | #333 |
| 7 | GPT-4o (2024-11-20) | 59.13 | #381 |
| 8 | MiniMax-M2 | 56.47 | #317 |
| 9 | DeepSeek R1 0528 | 54.36 | #222 |
| 10 | Qwen 3 Coder Plus | 46.27 | #338 |
| 11 | DeepSeek V3.1 (Thinking) | 41.12 | #264 (DeepSeek V3.1) |
| 12 | DeepSeek V3.1 | 40.65 | #264 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=webcoderbench-visual-experience · How It Works · Data refreshed daily, snapshot 2026-10-20.