Benchmark Almanac

DesignArena Multimodal to Video

Seedance 2.5 1,370 tops its leaderboard MiniMax H3 1,305 leads the models we track
Not in model cards

Video models make a video from a mix of text, images and other inputs. 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
18
Entries on its own leaderboard, 6 models we track

Seedance 2.5 tops the leaderboard with 1,370

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 1,305.

18 entries. Scroll sideways for the rest. Bars start at 900.

  1. 1,370 Seedance 2.5
  2. 1,334 Seedance 2.0 Fast
  3. 1,333 Seedance 2.0
  4. 1,305 MiniMax H3
  5. 1,284 Seedance 2.0 Mini
  6. 1,193 HeyGen Video 1.0
  7. 1,186 HappyHorse 1.0
  8. 1,160 Happy Horse 1.1
  9. 1,153 Wan 2.7
  10. 1,125 Gemini Omni 1.1 Flash
  11. 1,124 Kling O1 Edit
  12. 1,117 Gemini Omni 1.2 Flash
  13. 1,100 Gemini Omni Flash
  14. 1,006 Lucy 2.5
  15. 997 Gen-4 Aleph
  16. 997 Wan v2.6
  17. 988 LTX-2 19B
  18. 959 Wan Vace

Bars start at 900.

  1. 1 Seedance 2.5 1,370
  2. 2 Seedance 2.0 Fast 1,334
  3. 3 Seedance 2.0 1,333
  4. 4 MiniMax H3 1,305
  5. 5 Seedance 2.0 Mini 1,284
  6. 6 HeyGen Video 1.0 1080p, ultraquality 1,193
  7. 7 HappyHorse 1.0 1,186
  8. 8 Happy Horse 1.1 1,160
  9. 9 Wan 2.7 1,153
  10. 10 Gemini Omni 1.1 Flash 1,125
  11. 11 Kling O1 Edit 1,124
  12. 12 Gemini Omni 1.2 Flash 1,117
  13. 13 Gemini Omni Flash 1,100
  14. 14 Lucy 2.5 1,006
  15. 15 Gen-4 Aleph 997
  16. 16 Wan v2.6 997
  17. 17 LTX-2 19B 988
  18. 18 Wan Vace 959

Source: Official DesignArena Multimodal to Video leaderboard

#EntryScoreSettingDateSource
1 Seedance 2.5 1,370 Official leaderboard
2 Seedance 2.0 Fast 1,334 Official leaderboard
3 Seedance 2.0 1,333 Official leaderboard
4 MiniMax H3 1,305 Official leaderboard
5 Seedance 2.0 Mini 1,284 Official leaderboard
6 HeyGen Video 1.0 1,193 1080p, ultraquality Official leaderboard
7 HappyHorse 1.0 1,186 Official leaderboard
8 Happy Horse 1.1 1,160 Official leaderboard
9 Wan 2.7 1,153 Official leaderboard
10 Gemini Omni 1.1 Flash 1,125 Official leaderboard
11 Kling O1 Edit 1,124 Official leaderboard
12 Gemini Omni 1.2 Flash 1,117 Nov 2026 Official leaderboard
13 Gemini Omni Flash 1,100 Official leaderboard
14 Lucy 2.5 1,006 Official leaderboard
15 Gen-4 Aleph 997 Official leaderboard
16 Wan v2.6 997 Official leaderboard
17 LTX-2 19B 988 Official leaderboard
18 Wan Vace 959 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
1,117 2027 1,117 2027

Source: Official DesignArena Multimodal 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 233 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%