xAI
Grok Imagine Video 1.5
Grok Imagine Video 1.5 is xAI's newer video model, made to direct subject and camera motion from a prompt, with an optional start image to set the scene. It runs at 720p through OpenRouter for 215 credits a second, for 5, 10 or 15 seconds in five aspect ratios. A start frame adds 16 credits per clip.
Grok Imagine Video 1.5 is made by xAI. Prompt Generator is independent and not affiliated with xAI; we run the model through OpenRouter.
1,075 credits
- Model
- Grok Imagine Video 1.5 · xAI
- Cost
- 1,075 credits
Your results appear here when you press Generate.
Specs
- Vendor
- xAI
- Type
- Video
- Price
- 215 credits per second
- Aspect ratios
- 16:9, 9:16, 1:1, 4:3, 3:4
- Durations
- 5s, 10s, 15s
- Frames
- Start frame
- Input surcharge
- 16 credits per frame image
- Served via
- OpenRouter
Best for
- Directed subject and camera motion
- Clips up to 15 seconds
- Restyling a scene from one photo
Tips
- Draft on Grok Imagine Video 1.5 Lite (50 credits a second), then render the keeper here.
- Say what the camera does and what the subject does in separate sentences.
Questions
1What does a Grok Imagine Video 1.5 clip cost?
215 credits per second of video: a 5-second clip is 1,075 credits and a 15-second clip is 3,225. A start frame adds 16 credits, to cover the gateway's charge for an input image. If a run fails, the credits go back to your balance; runs stopped by the model's safety filter are refunded up to 10 times a day.
2Which lengths, sizes and frames does Grok Imagine Video 1.5 take?
Clips of 5, 10 or 15 seconds, in 16:9, 9:16, 1:1, 4:3 or 3:4. You can add one start frame (PNG, JPEG or WebP, up to 12 MB each) for image to video; without it Grok Imagine Video 1.5 works from text.
3Can I use what I make with Grok Imagine Video 1.5 in paid work?
Yes, when you made it on a paid plan or after buying credits. Results made only with the free sign-up credits are for personal, non-commercial use. Anything you upload must be yours to use, and people shown in it must have agreed.
Compare with all models.