# Veo 3.1 Reference to Video

> Video AI Model by Google — available on Layer

Reference-guided Veo 3.1: lock characters and objects across shots using up to three reference images.

Veo 3.1 Reference to Video is the reference-guided mode of Google's flagship Veo 3.1 video model. It accepts up to three reference images alongside a text prompt to keep characters, objects, and styles consistent across generated shots, while preserving Veo 3.1's photorealistic motion and native audio. Ideal for AI filmmakers and studios producing character-consistent narrative video at up to 4K resolution.

## Specifications

| Property | Value |
|----------|-------|
| Provider | Google |
| Category | Video |
| Price | 4.5 / second per second |
| Aspect Ratios | 16:9, 9:16 |
| Durations | 8s |

## Capabilities

- Audio Generation

---

[Try Veo 3.1 Reference to Video](https://next.app.layer.ai/me/session/new?baseModelId=VEO_3_1_REFERENCE) | [All Video Models](https://layer.ai/models/video) | [All Models](https://layer.ai/models)