Project AURIONAI Use-Cases Research & Innovation
LMArena (Arena)

LMArena Text-to-Video Arena

Arena score from pairwise human preference votes on text-to-video outputs

generative mediaactive

Score in Elo, LMArena Text-to-Video, updated 21 Sep 2026

  1. 1GoogleGemini Omni Flash 1.11,516 Eloindependent
  2. 2Google DeepMindGemini Omni Flash1,513 Eloindependent
  3. 3Black Forest LabsFLUX 3 Video1,493 Eloindependent
13scores
13models
13independent
0maintainer-run
0vendor-reported

Definition

Version
not in the notes
Category
generative media
Maintainer
LMArena (Arena)
Status
active
Home page
https://arena.ai/leaderboard/text-to-video

Sources

  1. Text-to-Video Arena: Best AI Video Generators arena.ai https://arena.ai/leaderboard/text-to-video
  2. Arena Leaderboard: Official AI Model Rankings & Benchmarks arena.ai https://arena.ai/leaderboard

Scores at a Glance

Score in Elo, LMArena Text-to-Video, updated 21 Sep 2026

Higher is better for this unit. Elo has no zero, so the axis starts at 1,200.

1GoogleGemini Omni Flash 1.1Googlegemini-omni-1.1-flash
1,516 Elo
independent2Google DeepMindGemini Omni FlashGoogle DeepMind, Google
1,513 Elo
independent3Black Forest LabsFLUX 3 VideoBlack Forest Labspreliminary
1,493 Elo
independent4SpaceXAI (xAI)Grok Imagine Video 1.5SpaceXAI (xAI)agent; preliminary
1,492 Elo
independent5ByteDanceSeedance 2.0ByteDance720p
1,479 Elo
independent6Alibaba (Qwen)Wan 3.0Alibaba (Qwen)
1,476 Elo
independent7ByteDanceSeedance 2.5ByteDance720p
1,474 Elo
independent8MiniMaxMiniMax H3 (Hailuo 3)MiniMax
1,460 Elo
independent9MetaMuse VideoMeta
1,456 Elo
independent10Alibaba (Qwen)HappyHorse 1.0Alibaba (Qwen)
1,427 Elo
independent11OpenAISora 2 ProOpenAI
1,368 Elo
independent12Google DeepMindVeo 3.1Google DeepMindveo-3.1-audio
1,364 Elo
independent13Alibaba (Qwen)Wan 2.7Alibaba (Qwen)wan2.7-t2v
1,337 Elo
independent
1,2001,3001,4001,5001,600
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.

Scores

13 of 13 scores shown lmarena-text-to-video-scores.csv
ModelScoreWho ran itSettingView or populationDateSource
GoogleGemini Omni Flash 1.11,516 Elo ±15independentgemini-omni-1.1-flashrank 1, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
Google DeepMindGemini Omni Flash1,513 Elo ±9independent–rank 2, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
Black Forest LabsFLUX 3 Video1,493 Elo ±17independentpreliminaryrank 3, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
SpaceXAI (xAI)Grok Imagine Video 1.51,492 Elo ±18independentagent; preliminaryrank 4, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
ByteDanceSeedance 2.01,479 Elo ±8independent720prank 5, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
Alibaba (Qwen)Wan 3.01,476 Elo ±13independent–rank 6, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
ByteDanceSeedance 2.51,474 Elo ±9independent720prank 7, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
MiniMaxMiniMax H3 (Hailuo 3)1,460 Elo ±9independent–rank 8, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
MetaMuse Video1,456 Elo ±15independent–rank 9, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
Alibaba (Qwen)HappyHorse 1.01,427 Elo ±13independent–rank 10, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
OpenAISora 2 Pro1,368 Elo ±7independent–rank 11, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
Google DeepMindVeo 3.11,364 Elo ±14independentveo-3.1-audiorank 12, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
Alibaba (Qwen)Wan 2.71,337 Elo ±7independentwan2.7-t2vrank 20, LMArena Text-to-Video, updated 21 Sep 202621 Sep 2026arena.ai
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "LMArena Text-to-Video Arena", https://aurion.semai.id/benchmarks/lmarena-text-to-video/, records of October 2026.