SuperCLUE-Mkt - Creative Production: leaderboard
SuperCLUE is the Chinese-language evaluation programme run by CLUE; its specialised boards each cover one professional domain and are graded by the organisers rather than self-reported. This board covers marketing: producing the creative itself.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Around January 2027. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 32B (Thinking) | 75.9 | #443 (Qwen 3 32B) |
| 2 | GLM-4.6 | 73.52 | #252 |
| 3 | DeepSeek V3.2 (Thinking) | 72.83 | #205 (DeepSeek V3.2) |
| 4 | Spark X1.5 | 72.72 | #390 |
| 5 | GPT-OSS-120B | 71.71 | #341 |
| 6 | Gemma 3 27B (IT) | 70.81 | #523 |
| 7 | GPT-5 Nano | 66.53 | #425 |
| 8 | DeepSeek R1 Distill Qwen 32B | 61.66 | #655 |
| 9 | GLM-4 32B (0414) | 59.4 | #533 |
| 10 | Llama 3.1 70B Instruct | 58.79 | #561 |
| 11 | InternLM3-8B-Instruct | 55.23 | #867 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-mkt-creative-production · How It Works · Data refreshed daily, snapshot 2026-10-20.