Create and Edit Video with Gemini Omni 1.1 Flash

Use Gemini Omni 1.1 Flash on Ezier to generate 3–10-second videos from text or images, guide first and last frames, preserve reference subjects, extend scenes, and refine edits with plain-language instructions.

Explore Gemini Omni 1.1 Flash video workflows

Move from a text brief to reference-led motion, then refine a clip without rebuilding the whole scene. Each prompt makes the subject, timing, camera, sound, and protected details explicit.

Prompt example

Create a 5-second photorealistic 16:9 shot looking straight down a weathered wooden pier that disappears into dense morning fog. Dozens of gray-and-white seagulls line both railings and gather on the damp boards; a few turn their heads, shuffle, flap, and lift briefly while the rest remain naturally spaced. Use a slow centered push-in, a level horizon, cool overcast light, calm gray water, and soft atmospheric depth. Generate distant gull calls, light wingbeats, faint water movement, and damp outdoor ambience. Keep the pier geometry, bird scale, fog density, and camera axis stable. No people, boats, text, logos, duplicate birds, warped railings, sudden weather changes, or cuts.

How to use Gemini Omni 1.1 Flash in Ezier

01
Select Gemini Omni 1.1 Flash
Choose the model, then confirm the available duration, aspect ratio, resolution, and credit estimate for the shot.
02
Assign each input a role
Start from text, add a first or last frame, attach subject references, or provide a source video for editing or extension.
03
Write the motion in order
Describe the action, camera path, timing, sound, and final state. Name every face, object, or composition detail that must stay consistent.
04
Generate, review, and refine
Check the result for continuity and instruction following, then request one focused edit or extend the strongest scene.

Plan generation, interpolation, and precise edits

Use text, images, and video as clearly assigned inputs. Direct the transition, preserve reference subjects, or request one focused change in natural language.

White exploration vehicle crossing a landscape that transitions from a desert observatory at sunrise to the same observatory under snow and stars.

Bridge first and last frames with planned motion

Provide a starting image and an ending image, then describe the path between them. Specify what changes over time and what must remain visually continuous.

Three consistent views of the same woman in a yellow raincoat carrying a navy backpack and silver camera at a coastal train station.

Carry reference subjects across the shot

Use reference images to define a person, object, or visual style. Repeat the identity details and tell the model whether each image is a reference or a literal frame.

Same dancer and museum pose shown twice, with only the dress material changing from burgundy velvet to reflective silver.

Change one detail and preserve the scene

Make edit instructions narrow and explicit. Name the single change, then protect the subject, pose, camera, background, timing, and sound that should stay untouched.

Gemini Omni 1.1 Flash output and control

Match the output to the job with a defined duration, format, resolution, and reference plan.

3–10s: Generated clip lengthBuild a compact shot or generate a continuation that lasts from three to ten seconds.
360p–4K: Resolution rangeDraft at 360p or 720p, then choose upscaled 1080p or 4K output for a higher-resolution delivery.
24 FPS: Output frame rateGenerated videos use a cinematic 24-frame-per-second cadence in 16:9 or 9:16 format.

Create from text or an image

Start with a detailed scene brief or animate a high-resolution image by naming the subject motion, camera movement, and environmental effects.

Connect two intentional keyframes

Use first-and-last-frame interpolation to plan where the sequence begins and ends while the model generates the movement between them.

Refine through conversation

Continue from a generated result and describe the next edit in plain language, protecting the details that should not change.

Extend a scene with reference media

Append a new three-to-ten-second continuation and introduce a referenced character or object while keeping the existing scene direction coherent.

Write stronger Gemini Omni 1.1 Flash prompts

Give every input a role, describe motion in sequence, and name the details that must remain unchanged.

Time each story beat

Use ranges such as [0–3s], [3–6s], and [6–10s] when the sequence needs several ordered actions.

Declare every reference role

State whether an image is the first frame, last frame, subject reference, object reference, or style reference.

Protect what should not change

For edits, name the one change first and follow it with the face, pose, camera path, background, and audio to preserve.

Put exclusions in the main prompt

Write constraints such as “no scene cuts,” “no dialogue,” or “no added objects” directly into the generation brief.

Ready to create?

Turn a clear brief into a controllable video sequence.

Define the scene, reference roles, camera path, sound, and protected details before you generate.

Gemini Omni 1.1 Flash FAQ

Answers about generation, reference media, interpolation, editing, duration, resolution, and prompt structure.

What is Gemini Omni 1.1 Flash?

Gemini Omni 1.1 Flash is Google's multimodal model for fast video generation, editing, interpolation, and extension. It accepts text, images, and video and produces high-resolution video with audio.

How long are Gemini Omni 1.1 Flash videos?

A generated clip or continuation can last from 3 to 10 seconds. Use timed prompt sections when several actions need to happen in a specific order.

Which resolutions does Gemini Omni 1.1 Flash support?

Google lists 360p and 720p generation, plus upscaled 1080p and 4K output. Videos are generated at 24 FPS in 16:9 or 9:16 format.

Can Gemini Omni 1.1 Flash animate an image?

Yes. Add a high-resolution source image and describe the subject movement, camera path, lighting change, and environmental motion you want to see.

Can Gemini Omni 1.1 Flash connect a first and last frame?

Yes. Provide two images as the starting and ending frames, then describe the transition, pacing, camera movement, and details that should remain consistent between them.

Can Gemini Omni 1.1 Flash edit or extend video?

Yes. Give the model a source video and a focused instruction to revise it or append a new 3–10-second continuation. Reference media can guide new subjects or objects in the extension.

How should I prompt Gemini Omni 1.1 Flash?

Describe the subject, action, setting, shot order, camera movement, lighting, sound, and final frame. Assign every uploaded image or video a clear role and state what must not change.

Next up

Create more with Gemini Omni 1.1 Flash

Try a nearby video model, open video tools, or prepare a stronger source image.

AI models

Compare models

Choose a better fit before you generate.

AI video tools

Turn images into motion

Animate a still or create a clip.

AI image tools

Create or edit images

Make, clean up, or upscale stills.

Omni Flash

Conversational editing

Use for short video generation and conversational edits.

Veo 3.1

Cinematic control

Use for cinematic shots and camera moves.

Seedance 2.5

Longer reference-led video

Use for longer scenes with richer reference direction.

Wan 3.0

Longer beta video generation

Use for text or start-image video up to 30 seconds.

LTX 2.5

4K audio-video workflow

Use for 1080p or 4K video with Fast, Pro, and generated-audio controls.

Kling 3.0

Action and motion

Use for action, motion, and story beats.