E-comIQ-18k: leaderboard
Metric: Acc@0.5 (%): share of the 1,000 E-comIQ-18k test posters (AI-generated Chinese e-commerce posters) whose predicted overall score falls within 0.5 of the expert-calibrated reference score on the 1-5 quality scale; higher is better. Source: arxiv.org. Saturation forecast: Around 2031. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4 | 33.3 | #169 |
| 2 | GPT-4o | 32.4 | #333 |
| 3 | Qwen 2.5 VL 7B Instruct | 29.3 | #643 |
| 4 | Claude Sonnet 4.5 | 28.4 | #138 |
| 5 | Gemini 2.5 Pro | 26.6 | #145 |
| 6 | Qwen 2.5 VL 72B Instruct | 18.6 | #364 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=e-comiq-18k · How It Works · Data refreshed daily, snapshot 2026-10-11.