# AI Video

AI Video turns a canvas image and a prompt into a clip. This pass uses Turbo MAX Image To Video. A smaller duration and resolution come back sooner.

## TL;DR: 6 steps

**AI Video** turns a picture already on the canvas, plus a text prompt, into a clip. On the bottom toolbar, open **AIGC**, then choose **AI Video**. The model menu includes Seedance 2.0, Veo 3.1 Fast, Seedance 2.5, and Kling Video V3 Pro. This walkthrough uses **Turbo MAX Image To Video**. Add the plate as the **First frame** with **Pick from canvas**, write the prompt, then run it. A shorter duration and a lower resolution come back sooner. This pass is 5 seconds at 480P.

The plate is a cat-garden scene.

![Cat-garden game scene used with AI Video](/images/ai-video/base_image.webp)

## Step 1: Open AI Video

On the bottom toolbar, open **AIGC**, then choose **AI Video**.

![AIGC menu with AI Video highlighted](/images/ai-video/step1_ai_video_generate.webp)

## Step 2: Choose the model and the size

Open the model menu. It lists Seedance 2.0, Veo 3.1 Fast, Seedance 2.5, Kling Video V3 Pro, and the Turbo MAX entries. This pass selects **Turbo MAX Image To Video**.

The control beside the model sets duration and resolution. A smaller size comes back sooner. This pass uses **5** seconds at **480P**.

![Model menu with Turbo MAX Image To Video selected](/images/ai-video/step2_select_model.webp)

## Step 3: Pick a first frame from the canvas

**First frame** is the starting picture. **End frame** can stay empty. Click the add control on **First frame**. The menu lists **Upload from device** and **Pick from canvas**. This walkthrough uses **Pick from canvas**.

![First-frame menu with Pick from canvas](/images/ai-video/step3_pick_from_canvas.webp)

## Step 4: Click the link icon on the image

Move the pointer onto the plate. A link icon appears on it. Click that icon. The image fills **First frame**.

![Link icon on the canvas image, and the filled First frame](/images/ai-video/step4_select_reference_image.webp)

## Step 5: Write the prompt

In the description box, write the clip. The text in this pass reads: `but featuring cute kittens, with an overall visual style inspired by *Kingdom Rush*. Please generate some audio tracks in a similar style that I can use as background music for the game.`

![Prompt filled in, with the garden plate as the first frame](/images/ai-video/step5_input_prompt.webp)

## Step 6: Review the clip on the canvas

Run the pass. The clip lands on the canvas with a play control. This one is 832×480, and the badge reads 5.2s.

![Generated clip on the canvas, 832 by 480, with a 5.2s badge](/images/ai-video/step6_video_output.webp)

The clip from this pass:

<figure class="video-embed" data-keep-place>
  <div class="video-embed-frame">
    <iframe
      src="https://www.youtube-nocookie.com/embed/mZ8dJYU6kds"
      title="Clip from this AI Video pass"
      allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
      allowfullscreen
    ></iframe>
  </div>
</figure>

## Walkthrough video

<figure class="video-embed">
  <div class="video-embed-frame">
    <iframe
      src="https://www.youtube-nocookie.com/embed/ewX5vIRdllc"
      title="VberAI Studio AI Video walkthrough"
      allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
      allowfullscreen
    ></iframe>
  </div>
</figure>
