Generate video with sound using Gemini Omni 1.2 Flash, the update Google released in November 2026. Pick any length from 3 to 15 seconds and 360p to 4K, in 16:9 or 9:16, from a prompt, frames or reference images.
The street cook was made with Gemini Omni 1.2 Flash at 1080p and the potter at 720p, both 6 seconds, from the prompt alone. Previews start muted. Tap Recreate to load the prompt and length into the generator above.
16:9 · 6sA street food cook at a night market tosses noodles in a flaming wok, then looks into the camera and says: "Two more minutes, it is worth the wait!" Sizzling oil, chatter of the crowd, warm neon light, handheld camera.
The cook tosses the noodles, looks into the lens and says the line from the prompt word for word, mouth in time, over sizzling oil and crowd noise. Google says 1.2 Flash makes dialogue more natural and lip-sync clearer.
9:16 · 6sVertical close-up: wet hands shape a clay bowl on a spinning pottery wheel in a sunlit studio, the camera slowly rises to reveal the finished rim, the soft hum of the wheel and quiet acoustic guitar music.
Ask for music and room sound and they arrive with the picture. This 9:16 pottery clip was made at 720p in about a minute; draft at 360p or 720p, then render the take you keep at 1080p or 4K.
| Resolutions | 360p, 720p, 1080p and 4K |
|---|---|
| Length | Any whole second from 3 to 15; 3 to 10 with first and last frames |
| Aspect ratio | 16:9 or 9:16 |
| Audio track | Speech, music and effects made with the picture; always on |
| Starting points | Text, a first frame, first and last frames, or up to 6 reference images |
Both versions run here, as does Google's Veo 3.1. What the 1.2 update changes:
| Change | 1.2 Flash | 1.1 Flash |
|---|---|---|
| Length range | 3 to 15 seconds | Up to 10 seconds |
| Length choices | Every whole second | 4, 6, 8 or 10 seconds |
| Images as references | Up to 6 | Up to 7 |
| Speech, per Google | More natural dialogue, clearer lip-sync | The earlier version |
Describe the action and the camera, then the audio: put spoken words in quotes and name any music or background noise.
Leave the frames empty for text to video, add a first frame (or first and last, up to 10 seconds), or use Reference with up to 6 images to keep a face or a product consistent.
Try ideas at 360p or 720p, then re-run the one you like at 1080p or 4K. The exact cost shows on the Generate button.
Presenters and characters who speak a line with matching lips.
Reference images keep packaging, colors and products consistent from shot to shot.
Up to 4K for the take you keep, after low-cost drafts at low resolution.
Pick the plan that matches your pace
Subscribers get +20% extra credits at the same price
≈ 18 videos5s · 480p each
Subscribers receive: 1,080
≈ 38 videos5s · 480p each
Subscribers receive: 2,280
≈ 110 videos5s · 480p each
Subscribers receive: 6,600
≈ 240 videos5s · 480p each
Subscribers receive: 14,400
Subscriptions renew automatically at the price shown until you cancel, and we'll tell you before any price change. Cancel anytime in Settings → Billing; you keep access until the paid period ends. Credit packs are one-time and expire 180 days after grant. Prices include any tax collected at checkout. Refunds & Billing · Terms · Privacy Policy · Contact Us · Pricing
aiMakeVideo is an independent AI video platform and is not affiliated with the model providers.
Gemini Omni 1.2 Flash is Google's video model with native audio, released on November 5, 2026 as an update to Gemini Omni 1.1 Flash. It makes 3 to 15 second clips with speech, music and sound effects.
Credits per second of video, more at higher resolutions: 360p is the cheapest and 4K the most. The Generate button shows the exact cost before you start, and credits are refunded if a run fails for a technical reason.
Any whole number of seconds from 3 to 15. With both a first and a last frame the limit is 10 seconds.
Yes, always. Speech, music and effects are made together with the picture, and there is no separate Sound switch. Write dialogue in quotes and describe the sounds you want.
Draft at 360p or 720p to find the shot, then render the final at 1080p or 4K. Our 6-second samples took about a minute at 720p and about a minute and a half at 1080p.
Not on aiMakeVideo. Even 3 seconds at 360p costs more than the free signup credits cover, so it needs a plan or credit pack. The free trial, where offered, makes one short Seedance 1 Pro Fast video with no account.
No. Gemini is Google's model, and the name belongs to its owner. aiMakeVideo is an independent platform, not affiliated with or endorsed by Google, where you can run Gemini Omni 1.2 Flash next to other video models.