C4STYLI: leaderboard
Large Language Models (LLMs) are increasingly deployed in diverse cultural contexts, yet their ability to master aesthetic stylistics, i.e., the strategic use of language to evoke cultural resonance, remains underexplored. We curate C4STYLI, a benchmark of highly stylized translated movie titles and advertising slogans from Hong Kong and the Chinese Mainland, to evaluate LLMs via the lens of behav
Metric: Accuracy (C4Styli-T) (self-reported). Source: benchmarklist.com. Saturation forecast: Around November 2026. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 79.9 | #454 |
| 2 | DeepSeek V3.2 (Thinking) | 74.5 | #205 (DeepSeek V3.2) |
| 3 | O4 Mini | 70.1 | #180 |
| 4 | Qwen 2.5 32B Instruct | 68.6 | #498 |
| 5 | GPT-4.1 | 68.5 | #248 |
| 6 | Qwen 3 14B | 68.5 | #537 |
| 7 | Qwen 3 32B | 68 | #443 |
| 8 | Qwen 3 8B | 66.1 | #674 |
| 9 | Qwen 2.5 14B Instruct | 64.8 | #647 |
| 10 | Llama 3 70B Instruct | 62.4 | #634 |
| 11 | Gemma 3 27B (IT) | 60.4 | #523 |
| 12 | Qwen 2.5 7B Instruct | 59.8 | #862 |
| 13 | Gemma 3 4B (IT) | 55.8 | #992 |
| 14 | Gemma 3 12B (IT) | 54.7 | #664 |
| 15 | Llama 3 8B Instruct | 53.5 | #1129 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=c4styli · How It Works · Data refreshed daily, snapshot 2026-10-20.