Every modality
Stills, video, audio, and text together.
Gemini Omni Video Generator
The full Gemini Omni engine — prompt-led video that reads image stills, video clips, and audio direction in one request.
AI 模型 36 點數
提示詞0/2000
請避免使用受版權保護的角色、標誌、音樂及敏感或違規內容，否則可能被攔截。
解析度
長寬比
時長 · 8s
批量生成1
預估費用：60 點數
你的作品將顯示在此
選擇模型、描述鏡頭，然後點擊生成。
為什麼選 Gemini Omni Video
為什麼創作者選擇 Gemini Omni Video
The full engine — every input type, one request.
720p to 4K
Draft fast, finish sharp.
Fast turnaround
The Omni engine built for iteration.
Free credits
Start without a card.
Gemini Omni Video guide
Gemini Omni Video — the full multimodal engine
Gemini Omni Video is the complete Gemini Omni engine in this studio — the model the Flash tiers are cut from. It reads a written shot alongside image stills, video clips, and audio direction, then renders coherent clips up to 4K.
Reference anything, create anything
Most video models read text or a single frame. Gemini Omni Video reads a stack: a prompt plus up to seven reference stills for cast, wardrobe, scene, and palette — with upstream hooks for video clips, character assets, and audio direction in the same request.
That makes it the model to reach for when the input is more than a sentence — a product that must stay accurate, a face that must stay consistent, or a scene defined by footage as much as words.
7 image references
Stills steer characters, scenes, and style.
Video + audio inputs
Clips and sound direction upstream.
To 4K
720p drafts through a 4K finish.
Video vs the Flash tiers
The Flash tiers are the speed cuts: 1.1 Flash adds a 360p draft mode and keyframes for cheap iteration; 1.2 Flash is the branded flagship for the finished shot. Gemini Omni Video is the full engine underneath — the widest input mix at the same clip lengths and resolutions.
A practical split: explore on 1.1 Flash at 360p, finish hero shots on 1.2 Flash, and reach for Gemini Omni Video when the shot needs video or audio conditioning alongside stills.
Directing it
Write the shot like a director — subject, action, camera, light — and name each reference’s job in the prompt. Draft at 720p to check the read, then move up in resolution only when the shot works.
Gemini Omni Video 使用方式
三步上手 Gemini Omni Video
Describe the shot
Subject, action, camera, light — name each reference’s role.
Set the format
720p, 1080p, or 4K — 16:9 or 9:16, 4–10 seconds.
Generate and iterate
Draft at 720p, finish at higher resolution.
Gemini Omni Video 適合誰
誰在用 Gemini Omni Video 做影片創作
Video creators
Reference-true shots and previs.
Marketers
Product and ad clips at volume.
E-commerce
Catalog video that stays on-product.
Indie creators
Shorts and loops that stay on-model.
更多模型
Gemini Omni Video 與全部模型都在同一個工作室
Switch engines without switching tools — Gemini Omni 1.2 Flash plus strong public image and video models, side by side in the Gemini Omni 1.2 Flash studio.
Text-to-videoReference-guided shotsUp to 7 image references720p–4K output16:9 & 9:164–10s clipsFast Flash draftsBatch runsGPT Image key visuals
Useful answers before you create
Gemini Omni Video 常見問題
Understand the model, prepare better references, and choose a practical approach to your next clip.
01What is Gemini Omni Video?
Google’s multimodal video engine — it fuses a text prompt with image, video, and audio references in one request to produce coherent clips up to 4K.
02What inputs does it accept?
A text prompt alone or with up to 7 reference images; the upstream engine also accepts video clips, character assets, and audio direction.
03How does it differ from the Flash tiers?
The Flash tiers (1.2 and 1.1) are the speed-optimized cuts for drafts and iteration; Gemini Omni Video is the full engine for the widest input mix.
04What lengths and resolutions are supported?
Clips run 4, 6, 8, or 10 seconds at 720p, 1080p, or 4K, in 16:9 landscape or 9:16 portrait.
05How much does a clip cost?
Billing scales with duration and resolution — the studio shows the exact credit estimate before you generate.
Make the next shot visible
Start your next clip on Gemini Omni 1.2 Flash
Bring a character still, a product photo, or a clear shot brief. Draft a take, refine the direction, and save the clip that fits your edit.