Benchmark Almanac

SuperCLUE Arena Image to Video

MiniMax H3 [768p] 1,205 tops its leaderboard
Not in model cards

Models animate images. The arena ranks systems from pairwise comparisons.

The arena reports median ratings and uncertainty intervals.

Made by
SuperCLUE
Kind
Video generation
Level
University
Scored
Ranked by matches
Answers
Open-ended
Tasks
All public
Measures
Preference
In and out
Images → Video
Interaction
Single turn
Items from
Live play or votes
Refresh
Live
Run by
A host
25
Entries on its own leaderboard, 0 models we track

MiniMax H3 [768p] tops the leaderboard with 1,205

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.

Bars start at 600.

  1. 1 MiniMax H3 [768p] 768p 1,205
  2. 2 Seedance 2.0 [720p] 720p 1,198
  3. 3 Seedance 2.5 [720p] 720p 1,173
  4. 4 Vidu Q3 [1080p] 1080p 1,155
  5. 5 可灵 3.0 [720p] 720p 1,146
  6. 6 PixVerse V6 [720p] 720p 1,115
  7. 7 可灵 2.5 Turbo [1080p] 1080p 1,105
  8. 8 HappyHorse-1.0 [1080p] 1080p 1,098
  9. 9 Seedance 1.0 [1080p] 1080p 1,074
  10. 10 Wan 2.7 [1080p] 1080p 1,074
  11. 11 可灵 O1(2025-12-01) [1080p] 1080p 1,069
  12. 12 Veo 3.1 [720p] 720p 1,067
  13. 13 Hailuo-2.3 [1080p] 1080p 1,055
  14. 14 Wan 2.5 Preview [1080p] 1080p 1,028
  15. 15 Vidu Q2 Turbo [720p] 720p 1,016
  16. 16 PixVerse V5.5 [1080p] 1080p 1,006
  17. 17 可灵 2.1 [720p] 720p 994
  18. 18 PixVerse V5 [1080p] 1080p 969
  19. 19 Wan 2.6(2025-12-16) [1080p] 1080p 952
  20. 20 Seedance 1.5 Pro [720p] 720p 869
  21. 21 PixVerse V4.5 833
  22. 22 Wan2.1-i2v-plus 803
  23. 23 Vidu Q1 727
  24. 24 I2V-01-Director 636
  25. 25 清影-AI生视频1.0 635

Source: Official SuperCLUE Arena Image to Video leaderboard

#EntryScore ±SettingDateSource
1 MiniMax H3 [768p] 1,205 1136.9 to 1266.9 768p Official leaderboard
2 Seedance 2.0 [720p] 1,198 1165.2 to 1234.9 720p Official leaderboard
3 Seedance 2.5 [720p] 1,173 1108.7 to 1232.6 720p Official leaderboard
4 Vidu Q3 [1080p] 1,155 ±24 1080p Official leaderboard
5 可灵 3.0 [720p] 1,146 1121.5 to 1170 720p Official leaderboard
6 PixVerse V6 [720p] 1,115 ±26.1 720p Official leaderboard
7 可灵 2.5 Turbo [1080p] 1,105 ±29.6 1080p Official leaderboard
8 HappyHorse-1.0 [1080p] 1,098 ±36.2 1080p Official leaderboard
9 Seedance 1.0 [1080p] 1,074 1044.2 to 1101.3 1080p Official leaderboard
10 Wan 2.7 [1080p] 1,074 ±33.1 1080p Official leaderboard
11 可灵 O1(2025-12-01) [1080p] 1,069 ±31.6 1080p Official leaderboard
12 Veo 3.1 [720p] 1,067 1038.9 to 1098.1 720p Official leaderboard
13 Hailuo-2.3 [1080p] 1,055 1025.4 to 1081.2 1080p Official leaderboard
14 Wan 2.5 Preview [1080p] 1,028 ±29.6 1080p Official leaderboard
15 Vidu Q2 Turbo [720p] 1,016 ±28.8 720p Official leaderboard
16 PixVerse V5.5 [1080p] 1,006 975.3 to 1038.4 1080p Official leaderboard
17 可灵 2.1 [720p] 994 968.7 to 1022.3 720p Official leaderboard
18 PixVerse V5 [1080p] 969 ±27.8 1080p Official leaderboard
19 Wan 2.6(2025-12-16) [1080p] 952 ±24.8 1080p Official leaderboard
20 Seedance 1.5 Pro [720p] 869 837.6 to 896.5 720p Official leaderboard
21 PixVerse V4.5 833 804.8 to 859.8 Official leaderboard
22 Wan2.1-i2v-plus 803 774.3 to 829.6 Official leaderboard
23 Vidu Q1 727 701.3 to 755.6 Official leaderboard
24 I2V-01-Director 636 606.5 to 664.8 Official leaderboard
25 清影-AI生视频1.0 635 ±32.4 Official 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

All 695 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

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (25 entries).
  • Pass Models scored 25 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as a rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on superclueai.com: 25 entries at the last read, none a model we track.
  • 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.

How we check