# Vidu 2 Reference

> A video generation model that accepts image and text input and returns video output.

## Key facts

- Model author: Z.ai
- Price: $0.40 per video
- Input modalities: image, text
- Output modalities: video
- Model IDs: `z-ai/vidu2-reference`
- Model page: https://www.realrelay.ai/model/z-ai/vidu2-reference

## Overview

Vidu 2 Reference is a video generation model reached through RealRelay's OpenAI-compatible API. It accepts image and text input and returns video output.

The vendor describes Vidu 2 as a model that strikes a balance between speed and quality.

Requests use the model ID z-ai/vidu2-reference.

## Capabilities

Vidu 2 Reference takes image and text input and produces video output.

The model is called with the model ID z-ai/vidu2-reference through RealRelay's OpenAI-compatible endpoint.

The vendor positions Vidu 2 as balancing generation speed with output quality.

## Pricing

Vidu 2 Reference is priced at $0.40 per video.

The price is stated per video rather than per token.

## Use Cases

Combine text and image input to produce video output.

Add video generation to an app through RealRelay's OpenAI-compatible endpoint.

Run video generation with image and text input at a per-video rate.

## FAQ

### What is Vidu 2 Reference?

Vidu 2 Reference is a video generation model accessed through RealRelay. It accepts image and text input and returns video output.

### How is Vidu 2 Reference priced?

The price is $0.40 per video.

### What input and output does Vidu 2 Reference support?

The model accepts image and text input and returns video output.

### How do I call Vidu 2 Reference through the OpenAI-compatible API?

Send a request to RealRelay's OpenAI-compatible endpoint with the model ID z-ai/vidu2-reference.

### What does the vendor say about Vidu 2 Reference?

The vendor describes Vidu 2 as striking a balance between speed and quality.