Video · Text → video · Image → video
Gemini Omni 1.1 Flash
Video from fast 360p drafts up to 4K — from text and up to 7 reference images
About the model
Gemini Omni 1.1 Flash is Google’s multimodal model for video generation. It’s built around a draft-to-final workflow: test an idea quickly in 360p, then produce the best take in 1080p or 4K.
The model responds well to directing cues — dolly zooms, snap zooms, push-ins, pull-backs, 360-degree camera orbits — and can draw on reference images to keep characters, scenes, and style consistent.
In our service, the model comes in two variants:
- Gemini Omni 1.1 Flash — 360p, 720p, or 1080p.
- Gemini Omni 1.1 Flash 4K — for final output at maximum quality.
Strengths
- Resolution from 360p for fast drafts up to 1080p and 4K for final output
- Up to 7 reference images — characters, scenes, style, or storyboards
- Expressive camera work — dolly zooms, snap zooms, push-ins, pull-backs, 360-degree orbits
- Prompts up to 20,000 characters for detailed scene descriptions
- Clip length of 4, 6, 8, or 10 seconds
Best for
- Ad and marketing clips
- Short story scenes and branded videos
- Iterating quickly on ideas in draft quality before the final version
- Vertical clips for social media
Limitations
- Maximum length is 10 seconds
- Aspect ratios: 16:9 and 9:16 only
- Up to 7 reference images, 20 MB each
Prompting tips
- Test the idea in 360p first — it's faster — then rerun the best take in 1080p or 4K
- Describe camera moves directly — "slow push-in", "orbit around the hero", "dolly zoom"
- Attach references of characters and settings to lock in their look
- For social media, pick the vertical 9:16 ratio