For frontier AI news

Image-to-Video Arena

Compare models according to their ability to generate videos based on the given images
Generate videos and vote in the Discord server

Last Updated

Dec 12, 2025

Total Votes

225,319

Total Models

25

Rank Spread
(Upper-Lower)
1
1◄─►2
1402±1612,439
Google
Proprietary
2
1◄─►2
1394±1412,266
Google
Proprietary
3
3◄─►5
1343±125,806
Alibaba
Proprietary
4
3◄─►5
1340±929,983
Google
Proprietary
5
3◄─►5
1331±938,234
Google
Proprietary
6
6◄─►8
1296±172,168
KlingAI
Proprietary
7
6◄─►10
1279±162,300
KlingAI
Proprietary
8
6◄─►8
1278±833,250
Bytedance
Proprietary
9
8◄─►12
1254±923,385
Google
Proprietary
10
9◄─►15
1252±923,350
Google
Proprietary
11
8◄─►18
1247±162,489
Shengshu
Proprietary
12
9◄─►16
Minimax
1246±89,535
MiniMax
Proprietary
13
10◄─►18
1236±732,103
KlingAI
Proprietary
14
11◄─►18
1232±1023,824
MiniMax
Proprietary
15
12◄─►18
1230±732,145
KlingAI
Proprietary
16
10◄─►18
1227±162,576
Shengshu
Proprietary
17
10◄─►18
Luma
1227±181,592
Luma AI
Proprietary
18
12◄─►18
1225±923,550
MiniMax
Proprietary
19
19◄─►20
1197±1024,544
MiniMax
Proprietary
20
19◄─►22
1185±733,009
Bytedance
Proprietary
21
20◄─►22
1171±929,665
Alibaba
Apache 2.0
22
20◄─►22
1170±1511,637
Google
Proprietary
23
23◄─►23
Luma
1109±1510,975
Luma AI
Proprietary
24
24◄─►24
1052±127,540
Runway
Proprietary
25
25◄─►25
997±129,485
Pika
Proprietary

Remove Style Control Leaderboard Plots

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)

Fraction of Model A Wins for All Non-tied A vs. B Battles

Confidence Intervals on Model Strength (via Bootstrapping)

Battle Count for Each Combination of Models (without Ties)