Video model

Gemini Omni Flash

Google's multimodal video model, with Gemini's sense of scenes and physics.

What it's best at.

  • Scenes where people and objects behave the way they would in real life.
  • Long, detailed prompts with several actions in the right order.
  • Turning a start image into a shot that keeps its look.

Specs.

Resolutions
720p · 1080p
Lengths
4 s · 6 s · 8 s · 10 s
Sound
Always on
Formats
16:9 · 9:16
Plans
Creator, Plus, Ultra and Teams

What it costs.

Credits per generation. You see the price before you start, and failed generations are refunded.

Resolutions4 s6 s8 s10 s
720p82 credits109 credits136 credits169 credits
1080p102 credits153 credits203 credits254 credits

Plus gives you 6,000 credits a month.

Prompts to try.

Copy one as it is, or make it yours.

  • A glass of water tips over on a wooden table, the water spreads and drips off the edge, slow motion, morning light.

  • A street magician makes a coin vanish in front of a surprised crowd, handheld close-up, city noise.

  • Vertical video: a paper plane glides through an open office, over the desks and out of a window into a blue sky.

Questions.

From 82 credits per video. You see the exact price before you generate, and if a generation fails, your credits come back.

Creator, Plus, Ultra and Teams. Every plan includes commercial use, with no watermark.

No. Gemini Omni Flash runs inside Aventada with your credits. No extra account, no API key.

Yes. Connect Aventada to Claude or any MCP app once, then ask for a generation in the chat. It uses your Aventada credits.

Other models.

All models

Use cases

Try Gemini Omni Flash.

Type it. Watch it.

Join the waitlist