Gemini Omni 1.2 Flash Image to Video

Gemini Omni 1.2 Flash Image to Video

What is Gemini Omni 1.2 Flash Image to Video?

Gemini Omni 1.2 Flash Image to Video, from Google, animates a single starting image into a 3-15 second clip with native, synchronized audio generated in the same pass. Supply a first frame — and an optional last frame when the ending matters — describe what should move and how it should sound, and Gemini Omni 1.2 Flash brings the still to life. It is the fast member of the Gemini Omni family, built for quick iteration from a static frame.

Key Features

• •First-frame animation, with an optional last frame (both frames cap the clip at 10 seconds)
• •Native synchronized audio on every clip — motion and sound in one pass
• •Legible on-screen text: titles, labels and numbers spelled as written
• •Timed, ordered beats from plain language or a timecode syntax
• •360p, 720p, 1080p and 4K output in 16:9 or 9:16, with a 360p draft-and-upscale workflow
• •Strong on stylised, non-photoreal looks: motion-graphics, claymation, low-poly and 3D
• •SynthID watermarking and C2PA Content Credentials for provenance

Best Use Cases

Gemini Omni 1.2 Flash is strongest on work that does not rely on photoreal people: animating infographics and data charts, adding motion and callouts to product shots, and stylised pieces in claymation, doodle, low-poly or 3D. The tested default example animates a flat "CITY RAINFALL" infographic — three bars grow in order to different heights and a summary tag slides in, each beat landing with its own sound. Draft at 360p, check the frames, then upscale the winner to 4K.

Prompt Tips and Output Quality

• •Pin the look in the first frame and describe only what should move.
• •Quote every on-screen word in single quotes and end with "every word spelled exactly as given, no other text".
• •Give beats in order (first, then, finally) or with a timecode, and write one sound per beat.
• •Use a locked-off camera and list what must stay still for clean layouts.
• •Supply a last frame when the ending matters, and check frames before upscaling.

How to Use Gemini Omni 1.2 Flash Image to Video

Gemini Omni 1.2 Flash Image to Video animates a starting image into a short clip with native sound. Give it a first frame, describe the motion and the audio, and keep everything else in the frame still.

Write the prompt

• •Pin the look in the first frame and describe only what should move; name a stylised medium (motion-graphics, claymation, low-poly) when you want a non-photoreal result.
• •Order the action as beats — "first..., then..., finally..." — or use a timecode syntax, and write one sound per beat.
• •For on-screen text, quote every word in single quotes and end with "every word spelled exactly as given, no other text".
• •Use a locked-off camera and list what must not move for clean, layout-driven scenes.

Choose the frames

• •first_frame_image sets the opening and the overall composition.
• •Add last_frame_image when the ending matters; with both frames the clip caps at 10 seconds.

Pick duration, resolution and ratio

• •duration runs 3-15 seconds.
• •resolution offers 360p, 720p, 1080p and 4k — draft the motion at 360p, review the frames, then re-run the winner at a higher resolution.
• •aspect_ratio is 16:9 for landscape or 9:16 for vertical and social.

Recommended Settings

A clean, tested default: a crisp first_frame_image, duration 5, resolution 720p, aspect_ratio 16:9. In the Segmind test, a flat "CITY RAINFALL" infographic animated cleanly — the bars grew in order to different heights, a "MM PER MONTH" tag slid in, and the title and labels stayed legible, each beat carrying its own sound. For finals, keep the same prompt and raise resolution. Roughly one instruction in four can be dropped, so check the frames and the text before you upscale.

Avoid photoreal people, acting and lip-sync — Gemini Omni 1.2 Flash is weaker there, and such prompts can trip content-safety filters.

Examples to try

Each example below ran end to end at the settings shown on its card — animated from a single starting frame with native audio generated in the same pass — so the first frame fixes the look and framing while the prompt's ordered beats carry the motion and the sound.

• •Animated infographic with legible on-screen text (the default example). Prompt: "Locked-off camera, nothing in the background moves. The three blue bars grow upward one after another into tall columns of different heights - first the 'JAN' bar to a short height, then the 'MAY' bar to a medium height, then the 'SEP' bar to the tallest. After the bars finish growing, a yellow rounded tag slides in at the top-right reading 'MM PER MONTH'. The deep teal background, the faint grid and the 'CITY RAINFALL' title stay perfectly still; only the bars and the tag move. Flat 2D vector style, no people, no photorealism. Sound: a short rising marimba note as each bar grows, then a light chime when the tag appears. Every word spelled exactly as given, no other text." At 720p the blue bars grew in order to different heights and the 'MM PER MONTH' tag slid in on a chime, the title and labels staying still. Quote every on-screen word in single quotes and close with "every word spelled exactly as given, no other text".
• •Pinned first and last frame. Prompt: "Claymation stop-motion style. A chubby brown modelling-clay treasure chest sits on flat sand-coloured clay ground against a plain pale-mint background. First the lid creaks and tilts, then it swings fully open, finally a faceted yellow clay gem rises into view inside the chest and gives off a small glow. The clip ends on the open chest with the yellow gem showing. Locked-off camera; the ground and the pale-mint background stay still. Sound: a wooden creak as the lid opens, then a bright chime when the gem appears. No people, no dialogue, no text." The clip opened on the closed chest and landed on the supplied open-chest frame, the lid creaking before a chime as the gem glowed. Add a last_frame_image in the same style and scale when the ending matters — both frames cap the clip at 10 seconds.
• •A stylised low-poly 3D look. Prompt: "Low-poly 3D isometric style. A small faceted wooden cabin with a stone chimney sits on a low-poly green hill under a flat pale-blue sky. First a yellow light switches on in the cabin window, then a thin ribbon of grey smoke rises from the chimney, finally a small lantern by the front door lights up. Locked-off isometric camera; the cabin, the hill and the sky stay still. Sound: a quiet evening wind throughout, a short click as the window light turns on, a faint crackle from the chimney, a gentle ding when the lantern lights. No people, no dialogue, no text." The window lit, low-poly smoke rose from the chimney and a porch lantern switched on in order, each beat carrying its own sound. Name the medium ("low-poly 3D isometric") and use a locked-off camera, listing what must stay still.

Further prompts this model is built for, as starting points rather than tested runs: "A doodle whiteboard explainer where three labelled arrows draw on one after another"; "A 9:16 claymation unboxing with a pop on each reveal"; "A timecode clip — '[0-3s] a low-poly car rolls in / [3-6s] it stops / [6-9s] the headlights blink on'."

What input does it take?

A starting image, plus an optional last frame.

Does it generate sound?

Yes — native synchronized audio on every clip.

How long can clips be?

3 to 15 seconds, or up to 10 seconds with both a first and last frame.

What resolutions and aspect ratios are supported?

360p, 720p, 1080p and 4K, in 16:9 or 9:16.

Can it render readable text?

Yes — quote the words in your prompt and they stay legible.

Is the output watermarked?

Yes — SynthID plus C2PA Content Credentials on every clip.