Veo 3.1: FAQ
1. What is Veo 3.1?
Veo 3.1 is Google's video generation model that produces clips with native, synchronized audio (dialogue, sound effects, ambience) from text, keyframes, or reference images. It supports text-to-video, image-to-video, and reference-based ("ingredient") generation in a single tool.
2. What resolution and duration does Veo 3.1 support?
Resolutions:
720p
1080p (16:9 only, and tied to the 8s duration)
Durations (clip lengths):
4s, 6s, or 8s
Available across all modes, including image-to-video
Must be 8s when using reference images
3. What if I want to extend my generation?
Veo 3.1 does not have a native extend feature; its clips top out at 8s. To lengthen a result, use the dedicated Extend feature (powered by Ray 3.14), which continues from the tail of an existing clip. You can also chain shots by feeding a Veo clip's final frame into a new generation as a start frame.
4. How does Veo 3.1 handle audio?
Audio is generated natively and synced automatically to the visuals, with no separate step. To direct it:
Dialogue: put spoken lines in quotes in the prompt (e.g.
A woman says, "We're here.").Sound effects & music: describe them explicitly (e.g.
thunder rumbling,soft synth pads).
5. Can you combine inputs with Veo 3.1 in one prompt?
Yes, with some rules:
Keyframes: a Start Frame can pair with an optional End Frame for a smooth transition.
Reference images: up to 3 ingredient images can be combined with a text prompt.
Not combinable: reference images cannot be used together with Start/End keyframes; pick one approach per generation. A text prompt is always required when using reference images.
6. What aspect ratios are available?
16:9 (landscape)
9:16 (portrait)
Auto-detected when keyframes are provided.
7. Are keyframes available with the Veo 3.1 model?
Yes. Veo 3.1 supports keyframe (image-to-video) generation:
Start Frame: the clip opens on this image.
End Frame: optional, requires a Start Frame, and creates a smooth transition into it.
Keyframes cannot be combined with reference images in the same generation.
Modes at a glance
Mode | What it does |
|---|---|
Text-to-video | Generates a clip from a written prompt alone, with native audio. |
Keyframe (image-to-video) | Starts on a Start Frame; optional End Frame creates a smooth transition. |
Reference (ingredients) | Uses up to 3 images to carry a subject or look into a new generation; needs a prompt. |
Within Reference mode, the Mode setting picks how references are applied:
Asset (Content): preserves a subject's appearance (person, product, character).
Style: transfers visual aesthetics (colors, lighting, texture).