How to Write Prompts for Veo 3.1
A practical prompt structure for short Veo 3.1 clips on Kensa: one subject, one action, a clear camera move, and how to iterate cheaply.
- Time to read
- 3 min read
- Model
- Veo 3.1
- Typical credits
- Typically 13 credits (8 s)
Quick Steps
Describe one shot
Write the prompt as a single shot: who or what is on screen, what they do, where it happens. An 8-second clip has room for one clear action, not a sequence of scenes.
Add the camera and the light
Say how the camera moves (slow push-in, tracking shot, static close-up) and what the light looks like (soft morning light, neon at night). These two details change the result more than adjectives do.
Pick the format before you generate
Choose the aspect ratio for where the clip will be used (9:16 for vertical social video, 16:9 for landscape). Every Veo 3.1 clip is 8 seconds long and costs 13 credits.
Change one thing per attempt
If the result is off, change one part of the prompt and generate again, so you can see what made the difference.
How to Write Prompts for Veo 3.1
Veo 3.1 is the fastest way to try an idea on Kensa: every 8-second clip costs a flat 13 credits, audio included. That makes it a good model for learning what a prompt does. This guide gives you a structure that works for short clips and a cheap way to improve it.
What you can set on Kensa
| Setting | Options |
|---|---|
| Length | 8 seconds |
| Aspect ratio | 16:9, 9:16 or 1:1 |
| Input | Text only, or text plus up to 2 images (a start frame, or a start and end frame) |
| Audio | Included |
| Cost | 13 credits per clip |
Your prompt text is sent to the model together with these settings. Prompts and uploaded images are checked by content moderation before generation; requests that break the usage policy are blocked.
A structure that works
Write one shot, in this order:
- Subject: who or what is on screen, with the details that matter ("a ceramic coffee mug with a blue glaze")
- Action: one thing that happens ("steam rises as a hand places it on a wooden table")
- Setting: where it happens ("a quiet café by a window")
- Camera: how the shot moves ("slow push-in, shallow depth of field")
- Light and style: the look ("soft morning light, warm colors, product commercial style")
Put together:
A ceramic coffee mug with a blue glaze, steam rising as a hand places it on a wooden table in a quiet café by a window. Slow push-in, shallow depth of field, soft morning light, warm colors, product commercial style.
Common problems and fixes
| What you see | Try |
|---|---|
| The clip tries to do too much and looks chaotic | Cut it down to one action; move extra ideas into separate clips |
| The camera wanders | Name one camera move, or ask for a static shot |
| The product changes shape or color | Start from a photo of the product (image-to-video) instead of describing it |
| The mood is flat | Describe the light and the time of day, not just "beautiful" or "cinematic" |
Iterate cheaply
Because every Veo 3.1 clip costs the same 13 credits, change one part of the prompt per attempt and only keep the wording that improves the result. When you have a prompt you like, reuse it as a base for variations: same camera and light, different subject.
When you need a clip longer than 8 seconds, try Seedance 2.0. To keep a specific product exactly as it looks, start from a photo with Image to Video.
Try it now
Open Text to Video, keep Veo 3.1 selected and try the example prompt above.