Benchmark Almanac

DesignArena Video Arena

Gemini Omni 1.2 Flash 1,369
Not in model cards

Users compare videos from anonymous video generation models for the same prompt and vote for the better one. Votes become Elo ratings.

Sibling boards cover image to video, multi-image to video, multimodal to video and video editing, each with its own record. One of 33 DesignArena boards. Its terms forbid automated reading, so the almanac imports no scores.

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
48
Entries on its own leaderboard, 15 models we track

Gemini Omni 1.2 Flash leads with 1,369

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.

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

  1. 1,369 Gemini Omni 1.2 Flash
  2. 1,357 Gemini Omni 1.1 Flash
  3. 1,338 Gemini Omni Flash
  4. 1,329 HeyGen Video 1.0
  5. 1,318 Wan 3.0
  6. 1,305 P-Video-2-Pro (Quality)
  7. 1,295 FLUX 3 Max
  8. 1,283 Pika H3 Flex
  9. 1,281 Flux 3 Video
  10. 1,278 Kling 4.0
  11. 1,271 MiniMax H3
  12. 1,247 Seedance 2.0
  13. 1,247 Seedance 2.5
  14. 1,246 Seedance 2.0 Fast
  15. 1,246 Seedance 2.0 Mini
  16. 1,216 Happy Horse 1.1
  17. 1,212 Grok Imagine
  18. 1,203 HappyHorse 1.0
  19. 1,174 Veo 3 Fast
  20. 1,168 Veo 3.1 Lite
  21. 1,161 Veo 3.1 Fast
  22. 1,160 Veo 3.1
  23. 1,158 LTX 2.5 Fast
  24. 1,157 Kling v3 Pro
  25. 1,157 Wan 2.7
  26. 1,153 LTX 2.5 Pro
  27. 1,144 Veo 3
  28. 1,140 P-Video 2
  29. 1,133 Sora 2
  30. 1,132 Sora 2 Pro
  31. 1,122 Kling O3 (Omni)
  32. 1,108 P-Video 2 Draft
  33. 1,093 Seedance 1.5 Pro
  34. 1,087 Ray 3.2
  35. 1,083 Kling v2.6 Pro
  36. 1,083 Ray 3.14
  37. 1,068 Kling v2.5 Turbo Pro
  38. 1,067 MiniMax Hailuo-2.3 (Pro)
  39. 1,060 MiniMax Hailuo-2.3 (Standard)
  40. 1,050 LTX-2.3
  41. 1,040 Glam.ai 1.0
  42. 1,031 P-Video (1080p)
  43. 1,011 Seedance 1.0 Pro Fast
  44. 984 Kandinsky 5.0 Pro
  45. 956 Kandinsky 5.0
  46. 937 LTX-2 19B
  47. 928 Ray2
  48. 914 Ray 2 Flash

Bars start at 900.

  1. 1 Gemini Omni 1.2 Flash 1,369
  2. 2 Gemini Omni 1.1 Flash 1,357
  3. 3 Gemini Omni Flash 1,338
  4. 4 HeyGen Video 1.0 1080p, ultraquality 1,329
  5. 5 Wan 3.0 1,318
  6. 6 P-Video-2-Pro (Quality) cost 1,305
  7. 7 FLUX 3 Max 1,295
  8. 8 Pika H3 Flex 1,283
  9. 9 Flux 3 Video 1,281
  10. 10 Kling 4.0 1,278
  11. 11 MiniMax H3 max 1,271
  12. 12 Seedance 2.0 1,247
  13. 13 Seedance 2.5 1,247
  14. 14 Seedance 2.0 Fast 1,246
  15. 15 Seedance 2.0 Mini 1,246
  16. 16 Happy Horse 1.1 1,216
  17. 17 Grok Imagine 1,212
  18. 18 HappyHorse 1.0 1,203
  19. 19 Veo 3 Fast 1,174
  20. 20 Veo 3.1 Lite 1,168
  21. 21 Veo 3.1 Fast 1,161
  22. 22 Veo 3.1 1,160
  23. 23 LTX 2.5 Fast 1,158
  24. 24 Kling v3 Pro 1,157
  25. 25 Wan 2.7 1,157
  26. 26 LTX 2.5 Pro 1,153
  27. 27 Veo 3 1,144
  28. 28 P-Video 2 1,140
  29. 29 Sora 2 1,133
  30. 30 Sora 2 Pro 1,132
  31. 31 Kling O3 (Omni) 1,122
  32. 32 P-Video 2 Draft 1,108
  33. 33 Seedance 1.5 Pro 1,093
  34. 34 Ray 3.2 1,087
  35. 35 Kling v2.6 Pro 1,083
  36. 36 Ray 3.14 1,083
  37. 37 Kling v2.5 Turbo Pro 1,068
  38. 38 MiniMax Hailuo-2.3 (Pro) 1,067
  39. 39 MiniMax Hailuo-2.3 (Standard) 1,060
  40. 40 LTX-2.3 1,050
  41. 41 Glam.ai 1.0 1,040
  42. 42 P-Video (1080p) 1,031
  43. 43 Seedance 1.0 Pro Fast 1,011
  44. 44 Kandinsky 5.0 Pro 984
  45. 45 Kandinsky 5.0 956
  46. 46 LTX-2 19B 937
  47. 47 Ray2 928
  48. 48 Ray 2 Flash 914

Source: Official DesignArena Video Arena leaderboard

#EntryScoreSettingDateSource
1 Gemini Omni 1.2 Flash 1,369 Nov 2026 Official leaderboard
2 Gemini Omni 1.1 Flash 1,357 Official leaderboard
3 Gemini Omni Flash 1,338 Official leaderboard
4 HeyGen Video 1.0 1,329 1080p, ultraquality Official leaderboard
5 Wan 3.0 1,318 Official leaderboard
6 P-Video-2-Pro (Quality) 1,305 cost Official leaderboard
7 FLUX 3 Max 1,295 Official leaderboard
8 Pika H3 Flex 1,283 Official leaderboard
9 Flux 3 Video 1,281 Official leaderboard
10 Kling 4.0 1,278 Official leaderboard
11 MiniMax H3 1,271 max Official leaderboard
12 Seedance 2.0 1,247 Official leaderboard
13 Seedance 2.5 1,247 Official leaderboard
14 Seedance 2.0 Fast 1,246 Official leaderboard
15 Seedance 2.0 Mini 1,246 Official leaderboard
16 Happy Horse 1.1 1,216 Official leaderboard
17 Grok Imagine 1,212 Official leaderboard
18 HappyHorse 1.0 1,203 Official leaderboard
19 Veo 3 Fast 1,174 Official leaderboard
20 Veo 3.1 Lite 1,168 Official leaderboard
21 Veo 3.1 Fast 1,161 Official leaderboard
22 Veo 3.1 1,160 Official leaderboard
23 LTX 2.5 Fast 1,158 Official leaderboard
24 Kling v3 Pro 1,157 Official leaderboard
25 Wan 2.7 1,157 Official leaderboard
26 LTX 2.5 Pro 1,153 Official leaderboard
27 Veo 3 1,144 Official leaderboard
28 P-Video 2 1,140 Official leaderboard
29 Sora 2 1,133 Official leaderboard
30 Sora 2 Pro 1,132 Official leaderboard
31 Kling O3 (Omni) 1,122 Official leaderboard
32 P-Video 2 Draft 1,108 Official leaderboard
33 Seedance 1.5 Pro 1,093 Official leaderboard
34 Ray 3.2 1,087 Official leaderboard
35 Kling v2.6 Pro 1,083 Official leaderboard
36 Ray 3.14 1,083 Official leaderboard
37 Kling v2.5 Turbo Pro 1,068 Official leaderboard
38 MiniMax Hailuo-2.3 (Pro) 1,067 Official leaderboard
39 MiniMax Hailuo-2.3 (Standard) 1,060 Official leaderboard
40 LTX-2.3 1,050 Mar 2026 Official leaderboard
41 Glam.ai 1.0 1,040 Official leaderboard
42 P-Video (1080p) 1,031 Official leaderboard
43 Seedance 1.0 Pro Fast 1,011 Official leaderboard
44 Kandinsky 5.0 Pro 984 Official leaderboard
45 Kandinsky 5.0 956 Official leaderboard
46 LTX-2 19B 937 Official leaderboard
47 Ray2 928 Official leaderboard
48 Ray 2 Flash 914 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
1,000 1,100 1,200 1,300 1,400 2026 2027 1,000 1,100 1,200 1,300 1,400 2026 2027

Source: Official DesignArena Video Arena 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%