AI video answer
How do I prompt Veo 3.1 image-to-video?
Let the input image define appearance and composition. Use the prompt to direct subject motion, camera movement, environmental reactions, sound, and the ending state.
Use the image as the visual anchor
The source frame already carries the character, product, setting, color, and composition. Repeat only the traits that must not drift, such as a face, logo, garment, or product shape.
Spend most of the prompt on what the still image cannot show: movement, timing, camera behavior, and sound.
Describe three kinds of motion
Direct subject motion, scene motion, and camera motion separately. For example, the runner tightens a lace, loose fabric moves in the breeze, and the camera slowly pushes closer.
Keep one movement dominant. Too many competing actions make a short image-to-video shot harder to control.
Match the current image-to-video limits
Google's current Veo 3.1 model page lists four-, six-, and eight-second generation, while reference-image-to-video uses eight seconds. It supports 16:9 and 9:16 output with sound generation.
Choose the output shape before writing the camera direction. Portrait and landscape frames give the subject and movement different amounts of room.
Related guide
Get the full context.
Veo 3.1 prompt guide, examples, and templates
Follow Google's five-part Veo 3.1 prompt formula: camera, subject, action, context, and style. Copy examples for dialogue, audio, and image-to-video.
Read the veo 3 1 prompt guide guideCurrent prices
Compare each video route.
| Model | Compared rate | Cre8vid price | 5 sec | 10 sec | Savings |
|---|---|---|---|---|---|
| MiniMax H3MiniMax | $0.065 | $0.029 | $0.145 | $0.29 | 55% less |
| Wan 3.0Alibaba | $0.050 | $0.025 | $0.125 | $0.25 | 50% less |
| Kling 3.0Kuaishou | $0.084 | $0.039 | $0.195 | $0.39 | 54% less |
| Seedance 2.5ByteDance | $0.144 | $0.069 | $0.35 | $0.69 | 52% less |
| Veo 3.1 Fast + audioGoogle | $0.100 | $0.049 | $0.24 | $0.49 | 51% less |
Source notes