Skip to content
IFAI

Model rankings

1022 models · 5 boards · 2 sources

Artificial AnalysisText-to-video arenaArenaText-to-video arena

Text-to-video arena · Anonymous head-to-head votes · Published September 22, 2026

Best video
gemini-omni-1.1-flashGoogle1,516Score
Best open model
minimax-h3MiniMax1,460Score
Best from China
dreamina-seedance-2.0-720pByteDance1,479Score
Most voted
grok-imagine-video-720pxAI165,462Votes
Top three by Score
2gemini-omni-flashGoogle1,513
3flux-3-videoBlack Forest Labs1,493
Top three open models
1minimax-h3MiniMax · #8 overall1,460
2kandinsky-5.0-t2v-proKandinsky · #35 overall1,175
3hunyuan-video-1.5Tencent · #36 overall1,169
This week's changes
HeyGen Video 1Video · #3New
Kimi K3.1 ProIntelligence · #4New
Gemini 4 Carbon (High)Coding · #4New
GPT-6.1 Terra (Max)Coding · #13New
MiniMax-M3.1-FlashIntelligence · #18New
RankChangeModelScore95% intervalVotes
1—gemini-omni-1.1-flashClosedGoogle1,516±151,784
2—gemini-omni-flashClosedGoogle1,513±926,576
3—flux-3-videoClosedBlack Forest Labs1,493±171,302
4—grok-imagine-video-1.5-agentClosedxAI1,492±181,218
5—dreamina-seedance-2.0-720pClosedByteDance1,479±856,318
6—wan3.0ClosedAlibaba1,476±132,598
7—dreamina-seedance-2.5-720pClosedByteDance1,474±98,001
8—minimax-h3OpenMiniMax1,460±911,417
9—muse-videoClosedMeta1,456±152,188
10—happyhorse-1.0ClosedAorizon1,427±1322,264
11—sora-2-proClosedOpenAI1,368±754,167
12—veo-3.1-audioClosedGoogle · 1 more variant1,364±1413,741
13—veo-3.1-fast-audioClosedGoogle · 1 more variant1,362±1039,484
14—veo-3-fast-audioClosedGoogle1,348±1125,258
15—sora-2ClosedOpenAI1,343±671,947
16—grok-imagine-video-720pClosedxAI1,342±6165,462
17—veo-3-audioClosedGoogle1,340±1319,002
18—wan2.7-t2vClosedWan1,337±727,626
19—wan2.6-t2vClosedAlibaba1,327±754,050
20—seedance-v1.5-proClosedByteDance1,256±776,171
21—veo-3ClosedGoogle1,254±1115,005
22—veo-3-fastClosedGoogle1,248±1215,283
23—wan2.5-t2v-previewClosedAlibaba1,245±936,378
24—pixverse-v5.6Closed1,240±1132,434
25—runway-gen-4.5ClosedRunway1,225±848,127
26—kling-2.5-turbo-1080pClosedKling1,220±172,111
27—kling-2.6-proClosedKling1,216±773,362
28—p-videoClosed1,207±167,079
29—ray-3ClosedLuma AI1,206±221,136
30—hailuo-2.3ClosedMiniMax1,206±689,138
31—kling-o1-proClosedKling1,205±271,194
32—hailuo-02-proClosedMiniMax1,198±139,399
33—seedance-v1-proClosedByteDance1,191±1112,164
34—hailuo-02-standardClosedMiniMax1,181±129,379
35—kandinsky-5.0-t2v-proOpenKandinsky1,175±212,019
36—hunyuan-video-1.5OpenTencent1,169±164,292
37—veo-2ClosedGoogle1,164±166,531
38—kling-v2.1-masterClosedKling1,163±1014,104
39—ltx-2-19bOpen1,154±875,596
40—wan-v2.2-a14bOpenAlibaba1,132±1510,458
41—seedance-v1-liteClosedByteDance1,113±1016,253
42—kandinsky-5.0-t2v-liteOpenKandinsky1,113±181,474
43—soraClosedOpenAI1,069±164,082
44—ray2ClosedLuma AI1,065±175,250
45—pika-v2.2ClosedPika1,009±155,763
46—mochi-v1OpenGenmo1,007±175,893

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi

Arena data comes from the leaderboard dataset it publishes on Hugging Face (CC BY 4.0). Only each board's overall leaderboard is used; scores and ranks are unchanged. dataset · CC BY 4.0

Artificial Analysis data comes from its API (artificialanalysis.ai). Scores are unchanged; ranks follow the scores. artificialanalysis.ai

Open or closed follows the license Arena gives; prices and Artificial Analysis intelligence show only where model names match, otherwise “—”.