Gemini Omni vs Veo 3.1: clips, prices and access compared

Gemini Omni vs Veo 3.1 is a choice between two Google video models that now overlap. Google released Gemini Omni 1.2 Flash on November 5, 2026. It makes clips up to 15 seconds and edits them in conversation. Veo 3.1 makes 4, 6 or 8 second clips, and its preview versions in the Gemini API were shut down on October 22, 2026. Veo 3.1 itself still runs on Google Cloud.

This comparison is based on Google's docs, pricing pages and announcements, and on published results. We did not generate video with either model for this article.

Gemini Omni vs Veo 3.1 at a glance

Sources: Google's Omni announcement, Google Cloud blog, Gemini API video docs and Veo 3.1 docs on Google Cloud, all read on November 5, 2026.

Clip length and resolution

Omni 1.2 Flash goes up to 15 seconds, up from 10 in Omni 1.1. Veo 3.1 stops at 8. If you need a single shot longer than 8 seconds, Omni covers it in one call. Both can extend a clip, so the gap matters less if you are willing to chain.

Resolution is closer. Veo 3.1 Generate outputs 720p, 1080p and 4K at 24 frames per second. Veo 3.1 Fast and Lite stop at 1080p, with Fast listing 4K pricing on the pricing page. Omni drafts at 360p or 720p and upscales the clips you keep. That draft-then-upscale flow is built to make iteration cheap.

How much each costs (as of November 5, 2026)

Omni is billed by token. Google's Gemini API pricing page lists gemini-omni-1.2-flash at $1.50 per million input tokens, $9.00 per million text output tokens and $12.00 per million video output tokens. A second of 720p video is 5,792 tokens, which Google's page turns into these approximate prices per second of output:

Google says 720p is 30% cheaper than Omni 1.1 Flash, which charged $17.50 per million video output tokens.

Veo 3.1 is priced on Google Cloud's generative AI pricing page in "count" units. The page does not define the unit in the text we read, so check it before you budget. As of November 5, 2026 it lists:

If the unit is a second of output, as in earlier Veo pricing, then Omni at about $0.07 per second of 720p video with audio undercuts Veo 3.1 Generate at $0.40, and sits between Veo 3.1 Fast and Lite. Treat that as a reading of the tables, not a quote from Google.

Where you can use each one

Omni 1.2 Flash is in Google AI Studio and the Gemini API (paid tier only), the Google Cloud console and enterprise API, Flow, Google Vids, Google Ads, and the Gemini app, where it is rolling out to Google AI Ultra subscribers. Google Vids offers 15-second generation at no additional cost. Audio references on Google Cloud are gated behind allowlisting.

Veo 3.1 is a Google Cloud model. Google's docs list the Generate and Fast models as generally available with a retirement date of May 4, 2027 or later, in us-central1 only. Lite is in preview. The docs flag the Fast model's status inconsistently, so confirm it before depending on it. Google's deprecations page names Omni 1.2 Flash as the replacement for the three Veo 3.1 preview models in the Gemini API. We did not check which Veo versions Flow uses.

What the sources do not tell us

Google does not publish a head-to-head comparison of Omni 1.2 Flash and Veo 3.1 on either page we read. We found no independent benchmark of the two on Google's pages, and quality claims such as better consistency and lip-sync come from Google. Google also lists known gaps for Omni: fast sports action, rapid script changes and complex physics. We did not compare Omni with Seedance 2.5 or other rivals.

Which one fits

For new work in the Gemini API, Omni is the only option, because the Veo 3.1 preview models are gone there. For Google Cloud teams with Veo pipelines, Veo 3.1 stays supported into 2027, and Lite is the cheapest listed option. Google says Omni outputs carry a SynthID watermark, and its docs say Veo on Google Cloud supports C2PA Content Credentials.

Watch the pricing pages. Google says Omni 1.1 Flash requests will route automatically to 1.2 Flash on December 3, 2026.