What it is
Gemini Omni Flash is Google's conversational video generation and editing model. The generally available version, gemini-omni-1.1-flash, was released on August 27, 2026, two months after the June 30 preview, and the preview endpoint was scheduled for deprecation on September 30. Google's docs now name it the model to use for video in the Gemini API, and the three Veo 3.1 preview models were shut down there on October 22, 2026; GA Veo models remain on Google's enterprise platform.
It makes 3 to 10 second clips at 24 frames per second, at 360p, 720p, 1080p or 4K (the last two by upscaling). It accepts text, images, audio and up to 10 seconds of video as input, and supports extension, first-and-last-frame interpolation and multi-turn edits through the Interactions API.
Our review
Omni Flash is a mid-pack generator whose real product is the edit loop, sold at a mid-table price. Artificial Analysis now ranks it tenth on its text-to-video board at 1130 Elo, behind Wan 3.0 (1175), MiniMax H3 and Seedance 2.5 (both 1158), but more than 160 points clear of every Veo 3.1 variant. At $9.12 a minute for 1080p it is a quarter of Seedance's price.
The reason to use it is revision. Because it runs on Gemini, you can ask for changes over several turns through the Interactions API and keep characters and scenes consistent, rather than regenerating from scratch. Version 1.1 added extension, first-and-last-frame interpolation and a resolution setting. Clips still stop at 10 seconds, and 1080p and 4K are upscaled from a lower native resolution rather than generated.
The Veo shutdown changes the calculation for existing Gemini API users. With the Veo 3.1 previews retired on October 22, anyone who built on them must either move to Omni Flash or to GA Veo on Google's enterprise platform. The move is not like-for-like: Google's model docs do not say whether Omni Flash generates a soundtrack, while Veo 3.1 is priced with and without audio on Runway.
There is no free tier, and the pricing page lists a per-second rate only for 720p. Runway publishes the full ladder, from $0.034 a second at 360p to $0.30 at 4K, which makes Runway the clearer way to budget higher resolutions even for teams on Google's API.
Where it earns its place
A backend developer whose app generated clips with veo-3.1-fast-generate-preview
Keep video generation working after the Veo previews were shut down on October 22
Switch to gemini-omni-1.1-flash on the same Gemini API key and move the call to the Interactions API. 5,000 seconds a month at 720p costs about $500; the alternative is migrating to GA Veo on Google's enterprise platform.
A social-media lead at a 15-person consumer brand
Turn one product still into a run of short vertical clips and tweak them without re-shooting
Animate the still, then ask for changes ('slower pan', 'swap the background to a kitchen') in follow-up turns. 500 ten-second clips at 720p is about $500 a month, roughly 40% of Seedance 2.5 on ModelArk at the same resolution.
A creative technologist at an agency already on Runway
Prototype storyboards cheaply before committing to a premium model for finals
Draft on Runway at 360p for $0.034 a second, so a 10-second animatic costs about 34 cents, then regenerate the chosen shots in a higher-ranked model such as Wan 3.0.
Before you commit
- Setup
- Requires a Gemini API key on a Cloud project with billing enabled; there is no free tier. Generation and multi-turn edits go through the Interactions API with a resolution setting (720p by default).
- Integrations
- Gemini API, Google AI Studio, the Gemini app and Runway's developer API.
- Data and privacy
- As a paid-only service, Google does not use prompts or outputs to improve its products, but logs them for an unstated period for abuse monitoring. The model docs do not say how long generated videos are stored.
- Cost at scale
- About $0.10 per second at 720p on the Gemini API, so 1,000 ten-second clips cost about $1,000; Runway charges $0.034 (360p), $0.10 (720p), $0.15 (1080p) and $0.30 (4K) per second, plus 1 credit per reference image and per second of input video.
- Limits
- 3 to 10 seconds at 24 fps; video input up to 10 seconds; 1080p and 4K upscaled. Rate limits, watermarking and rules on generating people are not stated on the model or video pages.
What's good
- Multi-turn conversational editing through the Interactions API
- About $0.10 per second at 720p, a quarter of Seedance 2.5 per minute
- Extension and first-and-last-frame interpolation since 1.1
What's not
- Clips limited to 3 to 10 seconds; video input capped at 10 seconds
- 1080p and 4K are upscaled, not native
- Tenth on Artificial Analysis's text-to-video board; audio output not documented
Best for
Developers on the Gemini API who want to generate and revise short clips through conversation.
Alternatives
Gemini Omni Flash 1.1: common questions
- Is Gemini Omni Flash 1.1 worth it?
- Gemini Omni Flash trades a few Elo points for the cheapest route to editing a clip by conversation, and with the Veo previews retired this week it is the only video model on the Gemini API price list. We rate it 4.3 out of 5.
- How much does Gemini Omni Flash 1.1 cost?
- $1.50 per 1M input tokens and $17.50 per 1M video output tokens, about $0.10 per second of 720p video; no free tier. Runway: $0.034 to $0.30 per second by resolution.
- What are the downsides of Gemini Omni Flash 1.1?
- Clips limited to 3 to 10 seconds; video input capped at 10 seconds; 1080p and 4K are upscaled, not native; tenth on Artificial Analysis's text-to-video board; audio output not documented.
- Who is Gemini Omni Flash 1.1 best for?
- Developers on the Gemini API who want to generate and revise short clips through conversation.
- How much does Gemini Omni Flash 1.1 cost for a team or in production?
- About $0.10 per second at 720p on the Gemini API, so 1,000 ten-second clips cost about $1,000; Runway charges $0.034 (360p), $0.10 (720p), $0.15 (1080p) and $0.30 (4K) per second, plus 1 credit per reference image and per second of input video.
- Is Gemini Omni Flash 1.1 safe for company data?
- As a paid-only service, Google does not use prompts or outputs to improve its products, but logs them for an unstated period for abuse monitoring. The model docs do not say how long generated videos are stored.
- What are the best Gemini Omni Flash 1.1 alternatives?
- The closest alternatives we have reviewed are Wan 3.0, Seedance 2.5, FLUX 3 and Kling 4.0.
Answers checked Oct 22, 2026.
Make Gemini Omni Flash 1.1? Show this rating on your site
