Skip to content
Katechon Arc
Start creating
Menu

ANSWER

What is Veo 3.1, and how do you use it?

Short answer

Veo 3.1 is Google's video model that generates 8-second clips at 720p, 1080p or 4K with native audio, guided by text and up to three reference images. You use it through Google's Gemini API and apps or through model API providers, and it is paid: $0.40 per second with audio on the Gemini API, with cheaper Fast and Lite tiers.

What Google says it does

Veo 3.1 is a model for generating 8-second videos (720p, 1080p or 4K) with natively generated audio. It accepts up to three reference images, supports portrait 9:16 video, video extension (limited to 720p) and generation from first and last frames. Source: Google AI for Developers, Veo 3.1 documentation (ai.google.dev/gemini-api/docs/veo), checked 2026-10-18

Veo 3.1 costs $0.40 per second with audio at 720p and 1080p and $0.60 at 4K on the Gemini API paid tier; Veo 3.1 Fast and Lite are cheaper; there is no free tier. Source: Google AI for Developers, Gemini API pricing (ai.google.dev/gemini-api/docs/pricing), checked 2026-10-18

Prices through a model API provider

Atlas Cloud list prices quoted 2026-10-17
TierResolutionUSD/s with audioUSD/s without8-second clip
Veo 3.1720p$0.400$0.200$3.20
Veo 3.1 Fast720p$0.096$0.080$0.77
Veo 3.1 Lite720p$0.050$0.050$0.40

Model prices on this page are Atlas Cloud list prices quoted 2026-10-17 with its free /calculate endpoint, as published in our AI video price index. Source: Atlas Cloud model catalog and /calculate list prices (atlascloud.ai/models), checked 2026-10-17

How to use it well

  1. Plan in 8-second beats. Each generation is one 8-second shot, so a scene is several clips cut together.
  2. Use reference images for anything that must stay the same: a character, a product, a set. You can give up to three.
  3. Use first and last frames when a shot must land on a specific composition, for example to match the next shot.
  4. Decide on audio up front. Through some providers Veo is cheaper without audio, which suits footage you will score yourself.
  5. Draft on Lite or Fast and re-render the approved shots on the full model.

What it is good for

  • Short ads and social clips where 8-second shots are the natural unit.
  • Shots that need native sound and a 1080p or 4K finish.
  • Less so for long, cheap series: at $0.400/s with audio it costs about 3 times Seedance 2.0 Mini per second.

We have not produced a film on Veo 3.1; our films were rendered on Seedance 2.0 models. The capabilities above are Google's own description.

Where Katechon Arc fits

If you are weighing Veo 3.1 for a multi-episode story, Arc's Seedance 2.0 models with reference sheets on every shot are a lower-cost option to price on the same shot list.

Arc cannot run Veo 3.1. Use Google's Gemini API, Google's apps or a provider that hosts Veo.

People also ask

Can you use Veo 3.1 for free?

Not on the Gemini API, which lists no free tier for Veo 3.1 as of 2026-10-18. Some consumer subscriptions include a limited number of Veo generations.

How long can a Veo 3.1 video be?

Google describes Veo 3.1 as generating 8-second videos; longer scenes are built from several clips or by extending a 720p video.

Related answers

Veo 3.1 vs Veo 3.1 Fast vs Veo 3.1 Lite: which tier should you use?On Google's Gemini API, Veo 3.1 costs $0.40 per second with audio at 720p and 1080p, Veo 3.1 Fast $0.10 at 720p and Veo 3.1 Lite $0.05 at…How much does Veo 3.1 cost?Google lists Veo 3.1 at $0.40 per second with audio at 720p and 1080p on the Gemini API, Veo 3.1 Fast at $0.10 (720p) and Veo 3.1 Lite at…Seedance 2.0 vs Kling 3.0 vs Veo 3.1: which should you use?For long series on a budget, Seedance 2.0 and its cheaper Fast and Mini tiers, which rendered every film we have released; for short…What is Kling 3.0, and what is it good for?Kling 3.0 is Kuaishou's video model for cinematic, talking clips of up to 15 seconds with native audio, multi-shot storytelling and…

Sources