SuperCLUE-AdsVU
Models interpret video advertisements.
SuperCLUE reports scores in a public workbook.
- Made by
- SuperCLUE
- Kind
- Video understanding
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Measures
- Capability
- Subject
- Media and communications
- Language
- Chinese
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- In views
- Vision
- Strongest model tested
-
Gemini 3.1 Pro #40 on the capability index
- Last model added
- June 2026
- Scores refreshed
- November 20th, 2026
Seed2.1 Pro leads with 89.1 among models released by June 2026 Seed2.1 Pro leads with 89.1 among models released by June 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in June 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1
Seed2.1 Pro high / API 89.1
- 2
Gemini 3.1 Pro high / API 88.5
- 3
Qwen3.7-Max API 86.9
- 4 Tec-Chi-VL-1.0-27B API 86.4
- 5 MiMo-V2.5 API 85.9
- 6 Kimi K2.6 API 83.1
- 7 GLM-5V-Turbo API 82.2
- 8
MiniMax-M3 API 67.2
Source: Official SuperCLUE-AdsVU leaderboard
- 1
Seed2.1 Pro high / API 89.1
- 2
Gemini 3.1 Pro high / API 88.5
- 3
Qwen3.7-Max API 86.9
- 4 MiMo-V2.5 API 85.9
- 5 Kimi K2.6 API 83.1
- 6 GLM-5V-Turbo API 82.2
- 7
MiniMax-M3 API 67.2
Source: Official SuperCLUE-AdsVU leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 89.1 | high / API | Jun 2026 | Official leaderboard |
| 2 | | 88.5 | high / API | Feb 2026 | Official leaderboard |
| 3 | | 86.9 | API | May 2026 | Official leaderboard |
| 4 | Tec-Chi-VL-1.0-27B | 86.4 | API | Official leaderboard | |
| 5 | MiMo-V2.5 | 85.9 | API | Apr 2026 | Official leaderboard |
| 6 | Kimi K2.6 | 83.1 | API | Apr 2026 | Official leaderboard |
| 7 | GLM-5V-Turbo | 82.2 | API | Apr 2026 | Official leaderboard |
| 8 | | 67.2 | API | Jun 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 89.1 | high / API | Jun 2026 | Official leaderboard |
| 2 | | 88.5 | high / API | Feb 2026 | Official leaderboard |
| 3 | | 86.9 | API | May 2026 | Official leaderboard |
| 4 | MiMo-V2.5 | 85.9 | API | Apr 2026 | Official leaderboard |
| 5 | Kimi K2.6 | 83.1 | API | Apr 2026 | Official leaderboard |
| 6 | GLM-5V-Turbo | 82.2 | API | Apr 2026 | Official leaderboard |
| 7 | | 67.2 | API | Jun 2026 | Official leaderboard |
Rankings by part
We read 22 parts from the benchmark’s own leaderboard in 4 groups: 创意元素识别, 创意内容理解, 创意策略分析 and 投放可行性判断. Each ranks models on one part of the test. Gemini 3.1 Pro leads 12 of them, and Seed2.1 Pro leads the other 10. Arrows show how far a model moved from its overall rank among the models we track.
创意元素识别
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 97.8
- 2
Seed2.1 Pro ↓1 97.0
- 3
Qwen3.7-Max 96.3
- 4 MiMo-V2.5 92.6
- 5 Kimi K2.6 91.2
- 6 GLM-5V-Turbo 87.5
- 7
MiniMax-M3 73.0
创意内容理解
7 models we track have a score here.
- 1
Seed2.1 Pro 81.8
- 2
Gemini 3.1 Pro 79.9
- 3
Qwen3.7-Max 78.1
- 4 GLM-5V-Turbo ↑2 77.9
- 5 MiMo-V2.5 ↓1 77.5
- 6 Kimi K2.6 ↓1 74.8
- 7
MiniMax-M3 61.6
创意策略分析
7 models we track have a score here.
- 1
Seed2.1 Pro 89.2
- 2
Gemini 3.1 Pro 89.2
- 3 MiMo-V2.5 ↑1 88.6
- 4
Qwen3.7-Max ↓1 87.8
- 5 Kimi K2.6 84.2
- 6 GLM-5V-Turbo 80.4
- 7
MiniMax-M3 67.0
投放可行性判断
7 models we track have a score here.
- 1
Seed2.1 Pro 91.6
- 2 GLM-5V-Turbo ↑4 90.3
- 3
Gemini 3.1 Pro ↓1 90.1
- 4 MiMo-V2.5 88.2
- 5
Qwen3.7-Max ↓2 87.4
- 6 Kimi K2.6 ↓1 86.8
- 7
MiniMax-M3 73.2
分数
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 97.8
- 2
Seed2.1 Pro ↓1 97.0
- 3
Qwen3.7-Max 96.3
- 4 MiMo-V2.5 92.6
- 5 Kimi K2.6 91.2
- 6 GLM-5V-Turbo 87.5
- 7
MiniMax-M3 73.0
品牌信息识别
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2
Qwen3.7-Max ↑1 100.0
- 3
Seed2.1 Pro ↓2 100.0
- 4 MiMo-V2.5 93.8
- 5 Kimi K2.6 87.5
- 6 GLM-5V-Turbo 81.3
- 7
MiniMax-M3 68.8
产品内容解析
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 88.9
- 2
Seed2.1 Pro ↓1 85.2
- 3 GLM-5V-Turbo ↑3 81.5
- 4 Kimi K2.6 ↑1 81.5
- 5 MiMo-V2.5 ↓1 81.5
- 6
Qwen3.7-Max ↓3 81.5
- 7
MiniMax-M3 59.3
行业标签识别
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2 Kimi K2.6 ↑3 100.0
- 3 MiMo-V2.5 ↑1 100.0
- 4
Qwen3.7-Max ↓1 100.0
- 5
Seed2.1 Pro ↓4 100.0
- 6 GLM-5V-Turbo 87.1
- 7
MiniMax-M3 61.3
达人画像识别
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2
Qwen3.7-Max ↑1 100.0
- 3
Seed2.1 Pro ↓2 100.0
- 4 Kimi K2.6 ↑1 93.9
- 5 GLM-5V-Turbo ↑1 87.9
- 6 MiMo-V2.5 ↓2 87.9
- 7
MiniMax-M3 75.8
调性风格归类
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2 GLM-5V-Turbo ↑4 100.0
- 3 MiMo-V2.5 ↑1 100.0
- 4
Qwen3.7-Max ↓1 100.0
- 5
MiniMax-M3 ↑2 100.0
- 6
Seed2.1 Pro ↓5 100.0
- 7 Kimi K2.6 ↓2 92.9
分数
7 models we track have a score here.
- 1
Seed2.1 Pro 81.8
- 2
Gemini 3.1 Pro 79.9
- 3
Qwen3.7-Max 78.1
- 4 GLM-5V-Turbo ↑2 77.9
- 5 MiMo-V2.5 ↓1 77.5
- 6 Kimi K2.6 ↓1 74.8
- 7
MiniMax-M3 61.6
叙事主旨提炼
7 models we track have a score here.
- 1
Seed2.1 Pro 73.8
- 2 GLM-5V-Turbo ↑4 71.8
- 3
Gemini 3.1 Pro ↓1 71.1
- 4
Qwen3.7-Max ↓1 71.0
- 5 Kimi K2.6 70.4
- 6 MiMo-V2.5 ↓2 69.7
- 7
MiniMax-M3 59.9
受众画像建模
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2 MiMo-V2.5 ↑2 100.0
- 3
Qwen3.7-Max 100.0
- 4
Seed2.1 Pro ↓3 100.0
- 5 GLM-5V-Turbo ↑1 92.9
- 6 Kimi K2.6 ↓1 85.7
- 7
MiniMax-M3 71.4
价值主张理解
7 models we track have a score here.
- 1
Seed2.1 Pro 71.5
- 2 GLM-5V-Turbo ↑4 69.2
- 3
Gemini 3.1 Pro ↓1 68.7
- 4 Kimi K2.6 ↑1 68.3
- 5
Qwen3.7-Max ↓2 63.1
- 6 MiMo-V2.5 ↓2 62.8
- 7
MiniMax-M3 53.4
分数
7 models we track have a score here.
- 1
Seed2.1 Pro 89.2
- 2
Gemini 3.1 Pro 89.2
- 3 MiMo-V2.5 ↑1 88.6
- 4
Qwen3.7-Max ↓1 87.8
- 5 Kimi K2.6 84.2
- 6 GLM-5V-Turbo 80.4
- 7
MiniMax-M3 67.0
黄金3秒钩子
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2 MiMo-V2.5 ↑2 100.0
- 3
Qwen3.7-Max 100.0
- 4
Seed2.1 Pro ↓3 100.0
- 5 GLM-5V-Turbo ↑1 92.5
- 6 Kimi K2.6 ↓1 84.9
- 7
MiniMax-M3 73.6
行动号召识别
7 models we track have a score here.
- 1
Seed2.1 Pro 70.9
- 2 Kimi K2.6 ↑3 67.7
- 3
Gemini 3.1 Pro ↓1 67.5
- 4 MiMo-V2.5 65.7
- 5
Qwen3.7-Max ↓2 63.4
- 6 GLM-5V-Turbo 55.5
- 7
MiniMax-M3 54.0
说服策略挖掘
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2 Kimi K2.6 ↑3 100.0
- 3 MiMo-V2.5 ↑1 100.0
- 4
Qwen3.7-Max ↓1 100.0
- 5
Seed2.1 Pro ↓4 96.7
- 6 GLM-5V-Turbo 93.3
- 7
MiniMax-M3 73.3
分数
7 models we track have a score here.
- 1
Seed2.1 Pro 91.6
- 2 GLM-5V-Turbo ↑4 90.3
- 3
Gemini 3.1 Pro ↓1 90.1
- 4 MiMo-V2.5 88.2
- 5
Qwen3.7-Max ↓2 87.4
- 6 Kimi K2.6 ↓1 86.8
- 7
MiniMax-M3 73.2
合规风控校验
7 models we track have a score here.
- 1
Seed2.1 Pro 74.9
- 2 Kimi K2.6 ↑3 71.8
- 3 GLM-5V-Turbo ↑3 71.0
- 4
Gemini 3.1 Pro ↓2 70.2
- 5 MiMo-V2.5 ↓1 64.6
- 6
Qwen3.7-Max ↓3 62.3
- 7
MiniMax-M3 57.4
场景适配性评估
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2 GLM-5V-Turbo ↑4 100.0
- 3 MiMo-V2.5 ↑1 100.0
- 4
Qwen3.7-Max ↓1 100.0
- 5
Seed2.1 Pro ↓4 100.0
- 6 Kimi K2.6 ↓1 88.7
- 7
MiniMax-M3 84.9
可信度评估
7 models we track have a score here.
- 1
Gemini 3.1 Pro ↑1 100.0
- 2 GLM-5V-Turbo ↑4 100.0
- 3 Kimi K2.6 ↑2 100.0
- 4 MiMo-V2.5 100.0
- 5
Qwen3.7-Max ↓2 100.0
- 6
Seed2.1 Pro ↓5 100.0
- 7
MiniMax-M3 77.1
2 models have held number one on the official leaderboard since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official SuperCLUE-AdsVU leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official SuperCLUE-AdsVU leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7%
- LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7%
- MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy.
Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy.
Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00%
- VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules.
Qwen3.5-Omni-Plus 93.1%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (8 entries).
- Pass Models scored 7 models we track, plus 1 other entry on its leaderboard.
- Pass Newest score The newest is from June 2026, for Seed2.1 Pro.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.1 Pro (#40).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on superclueai.com: 7 rows for models we track and 1 other entry at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.