AI Video Model · Google

Veo 3.1

Veo 3.1 is Google's Veo generation, emphasizing realistic motion, prompt adherence, and native audio: 4, 6, or 8 second clips (extendable) up to 4K, with first and last frames and up to three reference images. It is no longer Google's top-ranked video model — Gemini Omni Flash leads both blind-vote leaderboards at about a quarter of Veo 3.1 Standard's 720p price — but Veo 3.1 Lite is the cheapest audio-video rate we found.

At a glance (verified 2026-09-23)

Vendor
Google
Type
AI Video Model
Pricing
Gemini API, with audio: Standard $0.40/sec (720p/1080p; $0.60 at 4K), Fast $0.10/sec, Lite $0.05/sec (720p). Vertex AI without audio from $0.20 / $0.08 / $0.03.

Best for

Realism and prompt adherenceAudio-video clipsCinematic scenesAdvertising

How to use Veo 3.1 in Teamday

Not a native Teamday video route — Reel covers similar jobs today with Seedance 2 and 2.5, Kling 3 Pro, and Wan 2.7.

Frequently asked questions

Does Veo 3.1 generate native audio?

Yes — on the Gemini API audio is always on. Audio-free pricing exists only on Vertex AI (Standard $0.20/sec, Fast from $0.08, Lite from $0.03).

How much does Veo 3.1 cost per second?

With audio on the Gemini API: $0.40 per second for Veo 3.1 Standard at 720p or 1080p ($0.60 at 4K), $0.10 for Veo 3.1 Fast at 720p, and $0.05 for Veo 3.1 Lite at 720p. FAL and Replicate charge the same for Standard; Google direct is cheaper for Fast with audio ($0.10 vs $0.15).

Is Veo 3.1 still the best Google video model?

No. Gemini Omni Flash (gemini-omni-1.1-flash, generally available August 2026) ranked first on both the Artificial Analysis and Arena text-to-video leaderboards in September 2026 at about $0.10 per second at 720p. Veo 3.1 Lite remains a reasonable budget draft model.

Is Veo 3.1 available in Teamday?

Not currently — it is not exposed as a native video route. Teamday's integrated alternatives are Seedance 2 and 2.5, Kling 3 Pro (FAL), Wan 2.7, and Grok Imagine Video.

Related

Veo 3.1 is one of the models Teamday's AI employees already run — no API keys, no infrastructure.

Hire your first AI employee →