Gemini Omni
AI generationGoogleGoogle Gemini Omni Flash. Any-to-any multimodal video. Combine up to 7 images, an optional reference video, and reusable characters.
- Runs
- On our servers
- Typical time
- 30-180s
- Inputs
- 9
- Outputs
- 1
Inputs
| Port | Type | ID |
|---|---|---|
| Prompt | text | prompt-in |
| Image 1optional | image | image-in |
| Image 2optional | image | image-2 |
| Image 3optional | image | image-3 |
| Image 4optional | image | image-4 |
| Image 5optional | image | image-5 |
| Image 6optional | image | image-6 |
| Image 7optional | image | image-7 |
| Reference Videooptional | video | video-in |
Outputs
| Port | Type | ID |
|---|---|---|
| Video | video | video-out |
Settings
| Setting | Type | Default | Options |
|---|---|---|---|
| Duration | select | 4 | 4 seconds, 6 seconds, 8 seconds, 10 seconds |
| Aspect RatioOutput shape | select | 16:9 | 16:9 Landscape, 9:16 Portrait |
| ResolutionOutput resolution | select | 720p | 720p, 1080p, 4K |
| Seed0 = random. A fixed seed improves repeatability, but Omni results can still vary between runs. | number | 0 | 0 to 2147483647 |
Prompting
Describe the scene, camera move, or character action. Reference any uploaded images / reference video by what they show.
Maximum prompt length: 5000 characters.
What this connects into (18)
Computed from the port compatibility rules, so it stays correct as nodes change.
Node ID gemini-omni. Open it in Studio with Tab, then type the name.