Veo 3 on deep-imager is for motion that should feel photographed, not illustrated.
Still first with text to image, then Veo 3 for camera and light.
What is Veo 3 best for on deep-imager?
Cinematic b-roll, product motion, and realistic scene previews.
Use a different model if you want stylized anime motion. Veo 3 is the photoreal lane.
Should you storyboard with text to image first?
Yes — a still costs less to iterate than a clip, and image to image can lock framing.
Once the frame is right, switch to Veo 3 and add camera language: slow dolly, handheld, locked-off.
How do credits work for video vs stills?
Both draw from the same wallet: 5 free credits to start, then plans from $1/month.
Check how credits work for the current Veo 3 cost before a client preview.
- Draft stills first.
- One motion instruction per clip.
- Export MP4 when the queue completes.
Is image-to-video always on?
Check the model card in the AI image generator; support can differ by Veo version.
If image-to-video is listed, upload the still. If not, describe the scene in text and match the still by prompt.
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.