Pass IndexThe State of AISign in

grok-imagine-video

grok-imagine-video is xAI's original Grok Imagine video model, which works text-to-image then image-to-video under the hood: it generates a first frame from the prompt and animates it, producing clips up to 15 seconds at 480p or 720p in a range of aspect ratios, with an audio track by default.

text + image → audio + video · made by xAI

$0.05per second
xAI

Sold by 1 way

SellerLaneRate
xAI apistandard$3per minutedocs.x.ai · read 2026-08-25

Measured 1 standing

PlaceBoardMetricScore
20thof 32Artificial Analysis · Text to Video ArenaElo1062

About

grok-imagine-video — an audio and video model from xAI, sold by one company from $3 per minute, placed 20th of 32 on Artificial Analysis · Text to Video Arena.

It takes text and images and returns audio and video, with a context window of 1,024 tokens. It was published in January 2026. The catalogue files it under video. Only xAI sells it, at $3 per minute. It stands 20th of 32 on Artificial Analysis · Text to Video Arena.

Every current figure

Maker
xAI
Register
model
Takes
text + image
Returns
audio + video
Context
1,024 tokens
Published
January 2026
Licence
not read
Sellers
1
Price
$3 per minute — xAI
Boards
1
Best place
20th of 32 — Artificial Analysis · Text to Video Arena