# Gemini Omni Flash Reference

> Video AI Model by Google — available on Layer

Reference-guided video generation with up to 5 reference images for consistent characters and style.

Gemini Omni Flash Reference is the reference-guided variant of Google's Gemini Omni Flash video model. It accepts a text prompt alongside up to 5 reference images, which can be bound to roles inline in the prompt with tags like `<IMAGE_REF_0>`, letting creators lock characters and visual identity across a scene. Outputs are 720p clips of 3–10 seconds in 16:9 or 9:16 with synchronized native audio — ideal for character-consistent storytelling and branded content.

## Specifications

| Property | Value |
|----------|-------|
| Provider | Google |
| Category | Video |
| Price | 3.1 / megapixel second per megapixel_per_second |
| Aspect Ratios | 16:9, 9:16 |
| Durations | 3s, 4s, 5s, 6s, 7s, 8s, 9s, 10s |

---

[Try Gemini Omni Flash Reference](https://next.app.layer.ai/me/session/new?baseModelId=GEMINI_OMNI_FLASH_REFERENCE) | [All Video Models](https://layer.ai/models/video) | [All Models](https://layer.ai/models)