Gemini Omni AI Video Studio
Create and edit AI videos with Google Gemini Omni Flash. Build reusable characters and voice profiles, then compose multimodal videos from text, images, source clips, character IDs, and audio IDs.
Up to 7 image slots total (video uses 2, each character ID uses 1)
Estimated cost: $2.25
Result
Video generation usually takes 1–3 minutes. Please don't close this page
Transform Worlds Through Conversation
Official demo clips from Google I/O 2026 Gemini Omni showcase
Showcase
Official Google Gemini Omni Demos
These clips are from Google's I/O 2026 Gemini Omni showcase — conversational video editing, reference-guided style, and voice-driven generation in one model family.
Transform Worlds Through Conversation
Start from real footage and reshape the environment step by step — bubble structures become structural foam while the scene stays coherent across edits.
Why Gemini Omni
Google's Multimodal Video Creation Model
Gemini Omni Flash combines Gemini reasoning with generative video — text, image, video, character, and voice inputs in one API family.
Multimodal Video Generation
Generate 4–10 second clips at 720P, 1080P, or 4K from prompts plus up to 7 reference images, 1 source video, 1 audio ID, and 3 character IDs.
Conversational Video Editing
Upload source footage and transform scenes with natural language — change environments, actions, camera angles, and effects while keeping coherence.
Reusable Character Assets
Create stable character IDs from a portrait and description. Reuse them across Gemini Omni video generations for consistent identity.
Custom Voice Profiles
Generate reusable audio IDs from preset voices plus optional voice descriptions and sample dialogue.
Real-World Scene Logic
Gemini Omni connects visual creation with physics, narrative, and context — outputs feel intentional, not random.
Transparent Pay-As-You-Go Pricing
Pricing follows Kie.ai list rates at 3× markup. Video from 135 credits (4s 720P), character 20 credits, audio 15 credits.
How to Use
Create in 3 Steps
Use the Video, Character, and Audio tabs in one studio
Build Reusable Assets
Open the Character tab to create a character ID from a portrait, or the Audio tab to create a voice ID from a preset voice and description.
Compose Your Video Prompt
In the Video tab, describe the scene, add reference images or source video, and paste character/audio IDs as needed.
Generate & Download
Pick resolution, duration, and aspect ratio, then generate. Credits are charged upfront; failed video tasks are refunded automatically.
FAQ
Gemini Omni FAQ
Common questions about Gemini Omni on our platform
How is video pricing calculated?
Without source video: 720P/1080P costs 135/180/225/270 credits for 4/6/8/10 seconds; 4K costs 315/360/405/450 credits. With source video: 360 credits (720P/1080P) or 540 credits (4K) per generation. All prices are 3× Kie.ai list USD (100 credits = $1).
How much do Character and Audio cost?
Creating one reusable character ID costs 20 credits. Creating one reusable voice profile costs 15 credits. Both are fixed per-asset fees at 3× Kie upstream cost.
What inputs does Gemini Omni Video support?
Required prompt plus optional reference images (up to 7 slots), one trimmed source video (uses 2 slots), up to 3 character IDs (1 slot each), and one audio ID.
Can I edit existing footage?
Yes. Upload a source video in the Video tab. Output duration is determined automatically by the model; duration controls are disabled when video input is present.
What resolutions and durations are available?
Resolutions: 720p, 1080p, 4k. Durations: 4, 6, 8, or 10 seconds when no source video is uploaded. Aspect ratios: 16:9 or 9:16.
How long does generation take?
Most video jobs complete within 1–3 minutes. Character and audio asset creation is usually faster and returns IDs immediately.