Why is it not available inside a reel?

Upstream it expects @Image tags in the prompt to point at each reference, and our prompt compiler has no Grok dialect yet — it would emit prose with no tags. Whether the references still apply without tags is unverified, and reel is the wrong place to find out, since a character swap there is silent. It is fully available on the standalone video tool.

What resolution does it output?

480p or 720p. If you need 1080p or 4K, Kling V3 and Veo 3.1 go higher.

Grok Video or Hailuo 2.3?

Grok is cheaper and takes seven references against Hailuo's one. Hailuo reaches 1080p. For reference-driven work Grok, for a clean single-subject shot at higher resolution Hailuo.

Grok Imagine Video

Grok Imagine Video is xAI's video model. It accepts up to seven reference images — more than most models here — while costing less per clip than anything else we publish. For volume work where every shot does not need to be a hero shot, that combination is hard to argue with.

制作をはじめるショーケースを見る

モデル

スペック

モデル仕様

最大解像度

720p

最大の長さ

10s

速度

~40s per clip

クレジット単価

53 credits / 5s clip

できること

対応機能

AI Text to Video Generator

Reference to Video

仕組み

主な機能

Lowest Cost Per Clip

Cheaper than Hailuo 2.3, which had held that position. For a reel where most shots are establishing or cutaway, the saving compounds across the board.

Seven References

More than Kling V3 or Vidu Q3 accept. Useful when a shot has to carry several distinct subjects at once.

Verified on Both Providers

The reference field is named differently on each — reference_image_urls against image_urls — but both document the same semantics and the same cap, so a request carrying references can move between them.

ショーケース