SuperCLUE-I2V
Models generate videos from still images.
SuperCLUE reports scores in a public workbook.
- Made by
- SuperCLUE
- Kind
- Video generation
- Level
- University
- Scored
- Graded by people
- Answers
- Open-ended
- Measures
- Capability
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- In views
- Vision, Image, video and 3D generation
- Scores refreshed
- November 20th, 2026
MiniMax H3 leads with 71.5 MiniMax H3 leads with 71.5
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.
- 1
MiniMax H3 API 71.5
- 2
Gemini Omni 1.1 Flash API 67.4
- 3
Wan 3.0 API 66.6
- 4 Grok Imagine Video 1.5 API 65.8
- 5
Dreamina Seedance 2.5 API 64.6
- 6
Kling v3 Pro API 64.1
- 7 PixVerse V6 API 60.8
- 8
Vidu Q3 Pro API 60.4
Source: Official SuperCLUE-I2V leaderboard
- 1
MiniMax H3 API 71.5
- 2
Gemini Omni 1.1 Flash API 67.4
- 3
Wan 3.0 API 66.6
- 4 Grok Imagine Video 1.5 API 65.8
- 5
Dreamina Seedance 2.5 API 64.6
- 6
Kling v3 Pro API 64.1
- 7
Vidu Q3 Pro API 60.4
Source: Official SuperCLUE-I2V leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 71.5 | API | Official leaderboard | |
| 2 | | 67.4 | API | Official leaderboard | |
| 3 | | 66.6 | API | Official leaderboard | |
| 4 | Grok Imagine Video 1.5 | 65.8 | API | Official leaderboard | |
| 5 | | 64.6 | API | Official leaderboard | |
| 6 | | 64.1 | API | Official leaderboard | |
| 7 | PixVerse V6 | 60.8 | API | Official leaderboard | |
| 8 | | 60.4 | API | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 71.5 | API | Official leaderboard | |
| 2 | | 67.4 | API | Official leaderboard | |
| 3 | | 66.6 | API | Official leaderboard | |
| 4 | Grok Imagine Video 1.5 | 65.8 | API | Official leaderboard | |
| 5 | | 64.6 | API | Official leaderboard | |
| 6 | | 64.1 | API | Official leaderboard | |
| 7 | | 60.4 | API | Official leaderboard |
Rankings by part
We read 11 parts from the benchmark’s own leaderboard in 3 groups: 基础能力, 应用能力 and 回复率统计. Each ranks models on one part of the test. Wan 3.0 leads 4 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
基础能力
7 models we track have a score here.
- 1
Wan 3.0 ↑2 68.5
- 2
MiniMax H3 ↓1 67.1
- 3
Kling v3 Pro ↑3 65.7
- 4
Dreamina Seedance 2.5 ↑1 64.3
- 5 Grok Imagine Video 1.5 ↓1 64.2
- 6
Vidu Q3 Pro ↑1 61.2
- 7
Gemini Omni 1.1 Flash ↓5 60.1
应用能力
7 models we track have a score here.
- 1
MiniMax H3 75.8
- 2
Gemini Omni 1.1 Flash 74.6
- 3 Grok Imagine Video 1.5 ↑1 67.4
- 4
Dreamina Seedance 2.5 ↑1 64.9
- 5
Wan 3.0 ↓2 64.8
- 6
Kling v3 Pro 62.4
- 7
Vidu Q3 Pro 59.5
运动流畅性
7 models we track have a score here.
- 1
Gemini Omni 1.1 Flash ↑1 65.1
- 2
Wan 3.0 ↑1 64.2
- 3
MiniMax H3 ↓2 62.9
- 4
Kling v3 Pro ↑2 62.7
- 5 Grok Imagine Video 1.5 ↓1 61.7
- 6
Dreamina Seedance 2.5 ↓1 61.0
- 7
Vidu Q3 Pro 59.9
内容一致性
7 models we track have a score here.
- 1
MiniMax H3 67.3
- 2
Wan 3.0 ↑1 66.2
- 3
Kling v3 Pro ↑3 65.9
- 4 Grok Imagine Video 1.5 65.6
- 5
Dreamina Seedance 2.5 64.4
- 6
Vidu Q3 Pro ↑1 63.0
- 7
Gemini Omni 1.1 Flash ↓5 60.0
物理真实性
7 models we track have a score here.
- 1
Wan 3.0 ↑2 65.3
- 2 Grok Imagine Video 1.5 ↑2 63.3
- 3
Gemini Omni 1.1 Flash ↓1 62.7
- 4
Dreamina Seedance 2.5 ↑1 62.3
- 5
MiniMax H3 ↓4 61.8
- 6
Vidu Q3 Pro ↑1 57.1
- 7
Kling v3 Pro ↓1 57.0
长时序生成
7 models we track have a score here.
- 1
Wan 3.0 ↑2 69.9
- 2
MiniMax H3 ↓1 68.0
- 3
Kling v3 Pro ↑3 67.6
- 4
Dreamina Seedance 2.5 ↑1 65.3
- 5
Vidu Q3 Pro ↑2 59.2
- 6 Grok Imagine Video 1.5 ↓2 57.5
- 7
Gemini Omni 1.1 Flash ↓5 40.3
首尾帧控制
7 models we track have a score here.
- 1
Wan 3.0 ↑2 76.9
- 2
MiniMax H3 ↓1 75.5
- 3
Kling v3 Pro ↑3 75.5
- 4 Grok Imagine Video 1.5 72.8
- 5
Gemini Omni 1.1 Flash ↓3 72.6
- 6
Dreamina Seedance 2.5 ↓1 68.7
- 7
Vidu Q3 Pro 66.9
视觉风格创作
7 models we track have a score here.
- 1
Gemini Omni 1.1 Flash ↑1 68.4
- 2
MiniMax H3 ↓1 66.2
- 3
Kling v3 Pro ↑3 65.4
- 4
Vidu Q3 Pro ↑3 65.3
- 5 Grok Imagine Video 1.5 ↓1 65.2
- 6
Wan 3.0 ↓3 64.6
- 7
Dreamina Seedance 2.5 ↓2 61.3
广告营销
7 models we track have a score here.
- 1
Gemini Omni 1.1 Flash ↑1 75.3
- 2
MiniMax H3 ↓1 74.3
- 3 Grok Imagine Video 1.5 ↑1 71.7
- 4
Kling v3 Pro ↑2 71.5
- 5
Dreamina Seedance 2.5 70.9
- 6
Vidu Q3 Pro ↑1 70.7
- 7
Wan 3.0 ↓4 69.5
教学演示
7 models we track have a score here.
- 1
MiniMax H3 87.0
- 2
Gemini Omni 1.1 Flash 80.2
- 3 Grok Imagine Video 1.5 ↑1 65.3
- 4
Dreamina Seedance 2.5 ↑1 62.6
- 5
Wan 3.0 ↓2 60.1
- 6
Kling v3 Pro 50.3
- 7
Vidu Q3 Pro 42.5
回复率(%)
7 models we track have a score here.
- 1 Grok Imagine Video 1.5 ↑3 100.0
- 2
Kling v3 Pro ↑4 100.0
- 3
MiniMax H3 ↓2 100.0
- 4
Vidu Q3 Pro ↑3 100.0
- 5
Wan 3.0 ↓2 100.0
- 6
Dreamina Seedance 2.5 ↓1 97.3
- 7
Gemini Omni 1.1 Flash ↓5 96.0
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
No dated values yet.
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
7 of 8 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 2 pass 3 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.
- Note Newest score No scores yet.
- Note Frontier models None of its models are on the capabilities index.
- 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.