Benchmark Almanac

DesignArena Multi-Image to Video

Seedance 2.5 1,346 tops its leaderboard MiniMax H3 1,329 leads the models we track
Not in model cards

Video models compose one video from several reference images. Users compare two anonymous videos and vote for the better one.

One of 40 DesignArena boards. DesignArena ranks entries with Bradley-Terry ratings from user votes.

Kind
Video generation
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Last model added
November 2026
Scores refreshed
November 12th, 2026
25
Entries on its own leaderboard, 6 models we track

Seedance 2.5 tops the leaderboard with 1,346

Best score per entry from the benchmark's own leaderboard, read on November 12th, 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. The best model we track is MiniMax H3, at 1,329.

25 entries. Scroll sideways for the rest. Bars start at 600.

  1. 1,346 Seedance 2.5
  2. 1,329 MiniMax H3
  3. 1,324 Gemini Omni 1.2 Flash
  4. 1,308 Seedance 2.0
  5. 1,302 Seedance 2.0 Fast
  6. 1,286 Seedance 2.0 Mini
  7. 1,268 Kling 4.0
  8. 1,258 Gemini Omni 1.1 Flash
  9. 1,244 Happy Horse 1.1
  10. 1,226 Gemini Omni Flash
  11. 1,206 HappyHorse 1.0
  12. 1,199 Grok Imagine
  13. 1,198 Veo 3.1
  14. 1,187 Veo 3.1 Fast
  15. 1,181 Wan 2.7
  16. 1,151 Ray 3.14
  17. 1,101 Kling O1
  18. 1,038 Vidu Q2 Pro
  19. 999 Seedance 1.0 Lite
  20. 995 Ray 3.2
  21. 900 Kling v2.6 Pro
  22. 827 Kling O3 (Omni)
  23. 816 Seedance 1.5 Pro
  24. 814 Kling v2.5 Turbo Pro
  25. 774 Seedance 1.0 Pro Fast

Bars start at 600.

  1. 1 Seedance 2.5 1,346
  2. 2 MiniMax H3 1,329
  3. 3 Gemini Omni 1.2 Flash 1,324
  4. 4 Seedance 2.0 1,308
  5. 5 Seedance 2.0 Fast 1,302
  6. 6 Seedance 2.0 Mini 1,286
  7. 7 Kling 4.0 1,268
  8. 8 Gemini Omni 1.1 Flash 1,258
  9. 9 Happy Horse 1.1 1,244
  10. 10 Gemini Omni Flash 1,226
  11. 11 HappyHorse 1.0 1,206
  12. 12 Grok Imagine 1,199
  13. 13 Veo 3.1 1,198
  14. 14 Veo 3.1 Fast 1,187
  15. 15 Wan 2.7 1,181
  16. 16 Ray 3.14 1,151
  17. 17 Kling O1 1,101
  18. 18 Vidu Q2 Pro 1,038
  19. 19 Seedance 1.0 Lite 999
  20. 20 Ray 3.2 995
  21. 21 Kling v2.6 Pro 900
  22. 22 Kling O3 (Omni) 827
  23. 23 Seedance 1.5 Pro 816
  24. 24 Kling v2.5 Turbo Pro 814
  25. 25 Seedance 1.0 Pro Fast 774

Source: Official DesignArena Multi-Image to Video leaderboard

#EntryScoreSettingDateSource
1 Seedance 2.5 1,346 Official leaderboard
2 MiniMax H3 1,329 Official leaderboard
3 Gemini Omni 1.2 Flash 1,324 Nov 2026 Official leaderboard
4 Seedance 2.0 1,308 Official leaderboard
5 Seedance 2.0 Fast 1,302 Official leaderboard
6 Seedance 2.0 Mini 1,286 Official leaderboard
7 Kling 4.0 1,268 Official leaderboard
8 Gemini Omni 1.1 Flash 1,258 Official leaderboard
9 Happy Horse 1.1 1,244 Official leaderboard
10 Gemini Omni Flash 1,226 Official leaderboard
11 HappyHorse 1.0 1,206 Official leaderboard
12 Grok Imagine 1,199 Official leaderboard
13 Veo 3.1 1,198 Official leaderboard
14 Veo 3.1 Fast 1,187 Official leaderboard
15 Wan 2.7 1,181 Official leaderboard
16 Ray 3.14 1,151 Official leaderboard
17 Kling O1 1,101 Official leaderboard
18 Vidu Q2 Pro 1,038 Official leaderboard
19 Seedance 1.0 Lite 999 Official leaderboard
20 Ray 3.2 995 Official leaderboard
21 Kling v2.6 Pro 900 Official leaderboard
22 Kling O3 (Omni) 827 Official leaderboard
23 Seedance 1.5 Pro 816 Official leaderboard
24 Kling v2.5 Turbo Pro 814 Official leaderboard
25 Seedance 1.0 Pro Fast 774 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
1,324 2027 1,324 2027

Source: Official DesignArena Multi-Image to Video leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Audio and video

All 232 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%