AI video models

Gemini Omni Flash length, resolution, pricing and access

Google's multimodal video model you can edit by conversation. Every number here comes from Google's own pages, docs or pricing, with a link next to it. Where they don't say, we write "not published".

Specs and prices checked November 10, 2026

Gemini Omni Flash specs and pricing
Gemini Omni Flash is Google's newest video model and the replacement it names for the Veo previews. Version 1.2 (November 5, 2026) makes 3 to 15 second clips with native audio, extendable to 30 seconds, at 360p to 4K (upscaled). In the Gemini API it costs about $0.07 a second at 720p.

What are Gemini Omni Flash's specs?

SpecGemini Omni FlashSource
Made byGoogle
Model IDsgemini-omni-1.2-flash; gemini-omni-1.1-flash routes to 1.2 from December 3, 2026Gemini API: Omni Flash, Gemini API: deprecations
Released1.2 Flash: November 5, 2026. 1.1 Flash: August 27, 2026. Preview: June 30, 2026 (shut down October 22, 2026)Gemini API: deprecations
Max clip length3 to 15 seconds (default 10); extension adds 3 to 10 s per turn, up to 30 s totalGemini API: Omni Flash
Resolution360p, 720p (default), 1080p (upscaled), 4K (upscaled)Gemini API: Omni Flash
Aspect ratios16:9 (default), 9:16Gemini API: Omni Flash
Frame rateNot publishedGemini API: Omni Flash
AudioNative audio, generated by default. The API doesn't accept audio uploads; Google's launch post says the app takes voice referencesGemini API: Omni Flash, Google blog: Introducing Gemini Omni
InputsText, image (up to 6 reference images), video (references up to 3 s; edit uploaded videos up to 15 s); multi-turn editingGemini API: Omni Flash
License and weightsProprietary, hosted only. SynthID watermark on every videoGemini API: Omni Flash

From the developer's own pages, checked November 10, 2026. "Not published" means none of the developer's pages we could read states it.

How much does Gemini Omni Flash cost?

OptionPriceSource
Gemini API, 1.2 Flash$12.00 per 1M video output tokens; Google's estimate: $0.023/s (360p), ~$0.07/s (720p), $0.104/s (1080p), $0.209/s (4K)Gemini API: pricing
Gemini API, 1.1 Flash$17.50 per 1M video output tokens (about $0.10/s)Gemini API: pricing
Gemini API free tierNot availableGemini API: pricing
Via Runway's API (1.2)2 credits/s (360p), 7 (720p), 10 (1080p), 21 (4K), plus reference and input-video chargesRunway API: pricing

Google bills video by output tokens: 1,931 per second at 360p, 5,792 at 720p, 8,688 at 1080p, 17,376 at 4K. Its own example: about $0.70 for a 10-second 720p clip.

What that means per clip, computed from the per-second rates above (generation only, before retries):

OptionPer second5 seconds10 seconds30 seconds of footage
720p$0.07$0.35$0.70$2.10 (2 generations)
1080p (upscaled)$0.104$0.52$1.04$3.12 (2 generations)
4K (upscaled)$0.209$1.04$2.09$6.27 (2 generations)

Expect to pay for several takes per usable shot; these figures assume one.

Where can you use Gemini Omni Flash?

Video tools whose own pages name Gemini Omni Flash, checked November 10, 2026:

Only tools from our alternatives list, and only where the tool's own page names the model. A tool may use a different version or settings than the developer's API.

What are Gemini Omni Flash's limits?

What is Gemini Omni Flash good for?

Look elsewhere for: single generations over 15 seconds; aspect ratios other than 16:9 and 9:16. And for a launch or explainer video of your own software: generated clips can't show your real interface. That's a job for motion graphics tools, including nod, which builds the film from your product page.

How does Gemini Omni Flash compare with other AI video models?

ModelMakerMax clipMax resolutionAudioAPI from
Gemini Omni FlashGoogle15 s4K (upscaled)Yes$0.023/s (360p, Gemini API)
Veo 3.1Google8 s4KYes$0.03/s (Lite, 720p, video only)
Seedance 2.5ByteDance30 s1080pYes$0.103/s (480p, BytePlus)
Kling 4.0Kuaishou (Kling AI)30 s1080p (4K announced)Not stated$0.094/s (Kling 4.0 Flash, 720p, Runway API)
MiniMax H3MiniMax (Hailuo)15 s2KYes$0.025/s (H3-Lite, 480p)
Wan 3.0Alibaba30 s1080pNot stated$0.05/s (480p, Model Studio Singapore)
Luma Ray3.2Luma AI10 s1080p HDRNoCredits only
Sora 2OpenAI20 s1080p (Pro)YesShut down

Lowest per-second price each developer (or the named host) publishes, so tiers and resolutions differ: read each model's page before comparing. Full table on the AI video models hub.

Gemini Omni Flash FAQ

How long can a Gemini Omni Flash video be?
3 to 15 seconds (default 10); extension adds 3 to 10 s per turn, up to 30 s total.
Does Gemini Omni Flash generate audio?
Yes. Native audio, generated by default. The API doesn't accept audio uploads; Google's launch post says the app takes voice references.
How much does Gemini Omni Flash cost?
Gemini API, 1.2 Flash: $12.00 per 1M video output tokens; Google's estimate: $0.023/s (360p), ~$0.07/s (720p), $0.104/s (1080p), $0.209/s (4K). Gemini API, 1.1 Flash: $17.50 per 1M video output tokens (about $0.10/s). Google bills video by output tokens: 1,931 per second at 360p, 5,792 at 720p, 8,688 at 1080p, 17,376 at 4K. Its own example: about $0.70 for a 10-second 720p clip.
Is Gemini Omni Flash open source?
No. Proprietary, hosted only. SynthID watermark on every video.
Where can I use Gemini Omni Flash?
Gemini app and Google Flow (For Google AI Plus, Pro and Ultra subscribers, per Google's launch post); YouTube Shorts and YouTube Create ("at no cost", per Google's launch post (limits not stated)); Gemini API (Interactions API; no free tier); Runway API (gemini_omni_flash_1.2, 3 to 15 s, 360p to 4K). Tools whose own pages name it: Fliki and Renderforest.

More: All AI video models · Best AI video generators · Free AI video generators · How to make an AI video

Need a launch video, not a clip?

Paste your product page link. nod writes the script, designs the scenes and adds voiceover and music.