Native audio drafts
Generate a short clip where motion and the generated sound bed can be judged together.
AI Video Model
Gemini Omni AI video generator for audio-on prompt tests
Open AuraTuner with Gemini Omni Video when the first question is whether a prompt-led clip can carry generated audio and a short storyboard. Start with 720p and a narrow 6-second brief before trying longer or higher-resolution runs.
Starter setup
Start with this prompt
Mode
Text to video
Length
4, 6, 8, or 10 seconds
Audio
Generated audio enabled
Resolution
720p starter
Open AuraTuner with Gemini Omni Video when the first question is whether a prompt-led clip can carry generated audio and a short storyboard. Start with 720p and a narrow 6-second brief before trying longer or higher-resolution runs.
Use Gemini Omni when generated audio is part of the first test.
Start with 720p and 6 seconds before testing longer clips.
Use video-to-video only when an existing clip is the control input.
What you can do
These are the practical jobs this setup helps you finish faster.
Prompt-led scenes
Start from text when the idea, timing, and audio direction are still being explored.
Video remix entry
Switch to video-to-video when an existing clip should guide the next pass.
Start fast
Use this path before spending credits on variants or higher-quality runs.
Write the sound job
Name the ambient bed, product sound, or narration feel in the prompt.
Keep it short
Use 4 or 6 seconds for the first timing check before longer runs.
Save 4K for later
Use 720p or 1080p to test direction, then move to 4K after the idea works.
Short answer
Use Gemini Omni Video when it helps you finish the first useful draft.
Use Gemini Omni Video when you need a short prompt-led video with generated audio in the same first pass. Keep the prompt narrow so you can judge audio fit, subject stability, and timing before moving to longer or higher-resolution settings.
Related motion sample
- Sample type
- Product-motion reference
- Gemini setup
- Text to video, audio on
- First check
- Audio fit and motion timing
- Avoid
- Broad commercial prompts on run one
Best for
Use this model when these jobs match your first run.
Audio-on concepts
Short clips where ambient audio, timing, or spoken-style motion needs an early check.
Video remix tests
Video-to-video experiments when an existing source clip should guide the result.
Storyboard probes
4 to 10 second tests with one subject, one camera move, and one audio job.
First tests
Keep the first generation narrow and easy to grade.
Use 6 seconds
Long enough to hear timing, short enough to catch drift.
Name the audio job
Ask for ambient bed, product sound, or simple narration feel.
Keep one scene
Do not ask for a full ad, cutdowns, and scene changes at once.
Limits
Avoid wasting credits by checking these constraints.
Not audio upload
This is an audio-on video generation path, not a tool for syncing an uploaded audio track.
4K costs more
Use 720p or 1080p before testing 4K.
Video input cost
Video-to-video has a different cost profile from prompt-only runs.
Related entry pages
Gemini Omni Video FAQ
Does Gemini Omni Video in AuraTuner generate audio?
Yes. AuraTuner runs Gemini Omni Video with audio enabled, so the first test should include what kind of sound or timing you want to judge.
Can I upload audio into Gemini Omni Video?
No. Treat this as generated audio with video, not audio-to-video from an uploaded track.
What is the cheapest Gemini Omni starter setup?
The page starts with a 6-second 720p prompt-only run at 150 credits in AuraTuner. A 4-second 720p run is lower if you switch duration in Studio.
When should I use video-to-video?
Use video-to-video when an existing clip should guide the remix. Use text-to-video when the scene is still being discovered.
Can Gemini Omni replace my whole video editor?
No. Use it for the generated clip and audio direction. Keep final cuts, captions, music edits, and publishing review in your normal editor.