Sora 2 is the motion lane of the deep-imager AI image generator, not a separate site.
Keyframe in text to image, then prompt camera motion in Video mode.
Can you start a Sora 2 clip from a still?
Yes — generate a keyframe with text to image, then switch to Video mode and select Sora 2.
Describe subject, camera move, and lighting in one paragraph. Avoid conflicting motion.
Why keep stills and video on one AI image generator?
One prompt history and one credit wallet beats exporting a PNG to another app.
Image to image can tighten the keyframe before you spend a video credit.
How long does a clip take?
Plan 30 seconds–2 minutes depending on queue and model, same band as stills.
New users get 5 free credits to sample stills; video spends from the same pool, so test a still first.
What should you check before a paid video run?
Read how credits work and pricing from $1/month so a clip does not empty a Basic 10-credit plan by surprise.
Paid plans add commercial rights for client delivery. This is not an official OpenAI web app; it is deep-imager routing.
Start in text to image or image to image, pick a model such as GPT-4o image, then check pricing from $1/month and how credits work before a long run.