Google DeepMind's video model, known for polished lighting, close prompt adherence and native sound. Use it from a prompt, between a start and end frame, or guided by up to three reference images.
Try
From 300 credits
Up to 1080p
4s–8s
With audio
Made in Perfect Photo AI
1080p4s–8s2 ratiosAudio onfrom 300 cr
What Veo 3.1 is good at
Where Veo 3.1 stands out in the studio, and what it's worth reaching for it to make.
Follows the brief
Veo is good at doing what the prompt says: framing, lens, time of day and action land where you asked.
Start and end frames
Give it the first and last image and it fills in the motion between them, handy for transitions and product turns.
Reference images
Up to three references keep a subject or style consistent without using a start frame.
Fast variant
Veo 3.1 Fast costs less per clip at 720p, which suits testing prompts before a final 1080p render.
Is it the right model for this job?
A quick, honest read. Every model is in the same studio and uses the same credits, so switching costs nothing.
Pick Veo 3.1 when
You want a clean, high-end look with accurate framing
You need a controlled transition between two images
Yes. On Perfect Photo AI you pay per clip with credits instead of subscribing to Gemini or Flow. Credit packs are a one-time purchase and never expire.
Veo 3.1 renders at 720p or 1080p and gives the best quality. Veo 3.1 Fast is quicker and cheaper at 720p, which is useful for testing prompts.
4 or 8 seconds. 1080p and the end-frame and reference inputs are available on 8 second clips.
Yes. Use your image as the start frame, optionally add an end frame, and describe the motion.
Yes. Veo 3.1 generates audio with the video, including ambience, effects and dialogue you describe in the prompt.
16:9 for landscape and 9:16 for vertical shorts and Reels.
Yes. Videos you generate on Perfect Photo AI can be used for personal and commercial work, including ads and client projects, under our Terms of Service.