llm leaderboard

Best AI for video generation

Human preference over generated clips, from a prompt, from a still, and from an edit.

Aldena runs these models inside your team's projects. See what each one costs.

54 of 54 ranked models
Ranked models
rankmodelvendorcompositebenchmarkseloeloelo
1
Gemini Omni 1.2 Flashgoogle/gemini-omni-1.2-flash
Google76.13 of 3 benchmarks154015031396
2
Wan 3.0alibaba/wan-3.0
Alibaba74.93 of 3 benchmarks147614821424
3
Dreamina Seedance 2.5 720pbytedance/dreamina-seedance-2.5-720p
ByteDance74.83 of 3 benchmarks147914931417
4
Gemini Omni 1.1 Flashgoogle/gemini-omni-1.1-flash
Google72.82 of 3 benchmarks15131487
5
Minimax H3minimax/minimax-h3
MiniMax69.03 of 3 benchmarks145614931383
6
Gemini Omni Flashgoogle/gemini-omni-flash
Google68.33 of 3 benchmarks150914631378
7
Grok Imagine Video 1.5x-ai/grok-imagine-video-1.5
SpaceXAI68.02 of 3 benchmarks14881454
8
FLUX 3 Videoblack-forest-labs/flux-3-video
Black Forest Labs68.02 of 3 benchmarks14911452
9
Dreamina Seedance 2.0 720pbytedance/dreamina-seedance-2.0-720p
ByteDance65.43 of 3 benchmarks148314791370
10
Hidream O1 Video 1.0hidream/hidream-o1-video-1.0
HiDream61.31 of 3 benchmarks1455
11
Muse Videometa/muse-video
Meta60.61 of 3 benchmarks1457
12
Veo 3.1google/veo-3.1
Google60.42 of 3 benchmarks13631399
13
Veo 3.1 Fastgoogle/veo-3.1-fast
Google59.22 of 3 benchmarks13611386
14
HappyHorse 1.0alibaba/happyhorse-1.0
Alibaba58.73 of 3 benchmarks142914431305
15
Sora 2 Proopenai/sora-2-pro
OpenAI58.31 of 3 benchmarks1367
16
Wan 2.7alibaba/wan-2.7
Alibaba58.12 of 3 benchmarks13371428
17
Veo 3 Fastgoogle/veo-3-fast
Google56.32 of 3 benchmarks13471327
18
Sora 2openai/sora-2
OpenAI55.21 of 3 benchmarks1342
19
Veo 3google/veo-3
Google55.12 of 3 benchmarks13401331
20
Vidu Q3 Proshengshu/vidu-q3-pro
Shengshu54.91 of 3 benchmarks1363
21
Kling v3 Prokwaivgi/kling-v3-pro
Kling AI54.11 of 3 benchmarks1353
22
Wan2.6 T2Ialibaba/wan2.6-t2i
Alibaba52.22 of 3 benchmarks13241309
23
Wan2.5 I2I Previewalibaba/wan2.5-i2i-preview
Alibaba51.62 of 3 benchmarks12441322
24
Seedance 1.5 Probytedance/seedance-1-5-pro
ByteDance51.02 of 3 benchmarks12561307
25
Grok Imagine Videox-ai/grok-imagine-video
SpaceXAI50.93 of 3 benchmarks134114111250
26
Pixverse v5.6pixverse/pixverse-v5.6
PixVerse49.22 of 3 benchmarks12381299
27
Gen-4.5runway/gen-4.5
Runway49.01 of 3 benchmarks1223
28
Kling 2.6 Prokwaivgi/kling-2.6-pro
Kling AI46.92 of 3 benchmarks12161294
29
Kling 2.5 Turbo 1080pkwaivgi/kling-2.5-turbo-1080p
Kling AI46.92 of 3 benchmarks12201276
30
Vidu Q2 Turboshengshu/vidu-q2-turbo
Shengshu45.41 of 3 benchmarks1244
31
Hailuo 2.3minimax/hailuo-2.3
MiniMax44.02 of 3 benchmarks12051263
32
Kling O3 Prokwaivgi/kling-o3-pro
Kling AI43.51 of 3 benchmarks1263
33
Kling v2.1 Standardkwaivgi/kling-v2.1-standard
Kling AI43.01 of 3 benchmarks1230
34
Seedance v1 Probytedance/seedance-v1-pro
ByteDance42.82 of 3 benchmarks11921274
35
Kandinsky 5.0 T2V Prosber/kandinsky-5.0-t2v-pro
Sber AI42.01 of 3 benchmarks1175
36
Ray 3luma/ray-3
Luma AI41.52 of 3 benchmarks12061227
37
Hailuo 02 Prominimax/hailuo-02-pro
MiniMax41.02 of 3 benchmarks11991230
38
Vidu Q2 Proshengshu/vidu-q2-pro
Shengshu40.61 of 3 benchmarks1224
39
Hailuo 02 Fastminimax/hailuo-02-fast
MiniMax39.11 of 3 benchmarks1195
40
Kling v2.1 Masterkwaivgi/kling-v2.1-master
Kling AI38.12 of 3 benchmarks11631236
41
Hailuo 02 Standardminimax/hailuo-02-standard
MiniMax38.02 of 3 benchmarks11811225
42
Kandinsky 5.0 T2V Litesber/kandinsky-5.0-t2v-lite
Sber AI36.61 of 3 benchmarks1113
43
Kling O1 Prokwaivgi/kling-o1-pro
Kling AI36.32 of 3 benchmarks12011190
44
Soraopenai/sora
OpenAI35.81 of 3 benchmarks1069
45
Hunyuan Video 1.5tencent/hunyuan-video-1.5
Tencent35.72 of 3 benchmarks11681199
46
Runway Gen4 Turborunway/runway-gen4-turbo
Runway34.31 of 3 benchmarks1054
47
Mochi v1genmo/mochi-v1
Genmo33.51 of 3 benchmarks1007
48
Runway Gen4 Alephrunway/runway-gen4-aleph
Runway33.51 of 3 benchmarks1181
49
Veo 2google/veo-2
Google32.72 of 3 benchmarks11651167
50
Seedance v1 Litebytedance/seedance-v1-lite
ByteDance31.62 of 3 benchmarks11131186
51
Wan v2.2 A14balibaba/wan-v2.2-a14b
Alibaba31.62 of 3 benchmarks11321171
52
Ltx 2 19Blightricks/ltx-2-19b
Lightricks31.02 of 3 benchmarks11561161
53
Ray2luma/ray2
Luma AI27.52 of 3 benchmarks10661110
54
Pika v2.2pika/pika-v2.2
Pika25.72 of 3 benchmarks1009998
How this ranks

Every benchmark value becomes a percentile among the models that have it, so accuracy scores, Elo ratings and word error rates compare without hand-tuned scaling. Metrics where lower is better are inverted first. Raw values are never summed or averaged across benchmarks. A model's mean percentile is then shrunk toward the mean of the models that were broadly benchmarked, so a model tested twice cannot outrank a broadly tested one on two lucky results. Turning a data source off runs that same ranking code again in your browser over the sources you left on.

A model scored on fewer than 2 of the 3 ranked benchmarks in this category still ranks here, on the benchmarks it does have, and its row carries a partial coverage mark. On an equal score it sits under the model that earned the same number across more of the board.

Data sources

Turn a source off to drop every benchmark it feeds and re-rank the board in your browser. Your choice follows you across the leaderboard pages.

  • LMArenaCC BY 4.0

    Arena ratings by LMArena, from the public leaderboard dataset.