Video · guidance updated 4 Oct 2026
Google Veo / Gemini Omni (Gemini app & Flow) prompt guide
The Gemini app and Flow now use Gemini Omni Flash 1.1 (May 2026): 3–10 s clips extendable to 40 s, native audio, and conversational editing. Veo 3.1 remains in the API and Vertex (4/6/8 s, native audio, 16:9 or 9:16).
How to prompt Google Veo / Gemini Omni (Gemini app & Flow)
- Cover framing/camera, style, lighting, location and action; Omni fills small details from your overall intent.
- Use film camera terms: wide-angle tracking shot, low angle, static, push in, dolly zoom, aerial/drone, close-up.
- Put exact spoken words in quotation marks and name who says them.
- Describe sound effects and ambience explicitly, e.g. 'tires screeching, engine roaring, faint hum in background'.
- Ask for 'one continuous shot' or 'single unbroken scene' if you don't want cuts.
- For on-screen text, give the exact words, where they appear, and when.
- For timing, use timecodes such as [0-3s] … [3-6s] … (Omni).
- Give characters specific looks (age, hair, clothing), not generic labels.
- Use vivid sensory adjectives and a named style (claymation, film noir on 35mm, VHS).
- For edits, change one thing per turn and add 'Keep everything else the same.'
- With Omni, state exclusions plainly ('No dialogue', 'No music'); with Veo, use the separate negative-prompt field.
The shape of a good prompt
[Shot framing + camera motion] of [subject with specific details] [action] in [location]. [Style]. [Lighting/ambiance]. [Optional timecodes: [0-3s] … [3-6s] …]. Dialogue: [Character] says, "[line]". Audio: [SFX], [ambient], [music]. [Optional: One continuous shot. No …]
Avoid
- Dialogue too long for the clip length
- Generic characters ('a man')
- Leaving audio unspecified when sound matters
- Several unrelated edits in one follow-up turn
- Uploading or depicting real people you don't have rights to
Example
Request a red convertible driving along the coast
Drone shot following a classic red convertible along a winding coastal cliff road at golden hour, waves crashing on the rocks below. One continuous shot. Warm, low sunlight with long shadows; cinematic 35mm film look. The driver, a woman in her thirties with a white headscarf and sunglasses, glances at the sea and smiles. Audio: the engine roars as it accelerates, wind rushing, distant gulls, surf crashing. No music.
