Model rankings

1055 models · 5 boards · 2 sources

Week of March 9, 2026

April 20, 202627 changes April 13, 202637 changes April 6, 202641 changes March 30, 2026230 changes March 23, 202626 changes March 16, 202622 changes March 9, 202665 changes March 2, 202663 changes February 23, 202613 changes February 16, 202616 changes February 9, 202641 changes February 2, 20266 changes

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

Text-to-video arena · Anonymous head-to-head votes · Published March 6, 2026

Best video

veo-3.1-audio-1080pGoogle1,381Score

Best open model

kandinsky-5.0-t2v-proKandinsky1,179Score

Best from China

wan2.6-t2vAlibaba1,347Score

Most voted

kling-2.6-proKling38,740Votes

Top three by Score

1veo-3.1-audio-1080pGoogle1,381

2veo-3.1-fast-audio-1080pGoogle1,378

3sora-2-proOpenAI1,367

Top three open models

1kandinsky-5.0-t2v-proKandinsky · #24 overall1,179

2hunyuan-video-1.5Tencent · #25 overall1,171

3wan-v2.2-a14bAlibaba · #28 overall1,130

This week's changes

GPT-5.4 (xhigh)Intelligence · #2New

gpt-5.4-high (codex-harness)Coding · #5New

gpt-5.4-medium (codex-harness)Coding · #12New

Step 3.5 FlashIntelligence · #35New

Gemini 3.1 Flash-Lite PreviewIntelligence · #46New

All Open Closed

All regions US China

One row per model All variants

SortScore Votes

Rank	Change	Model	Score	95% interval	Votes
1	—	kandinsky-5.0-t2v-proOpenKandinsky	1,179	±21	1,886
2	—	hunyuan-video-1.5OpenTencent	1,171	±16	4,097
3	—	wan-v2.2-a14bOpenAlibaba	1,130	±15	11,158
4	—	ltx-2-19bOpenLightricks	1,122	±10	21,120
5	—	kandinsky-5.0-t2v-liteOpenKandinsky	1,112	±18	1,353
6	—	mochi-v1OpenGenmo	999	±16	6,676

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