Skip to main content

Generating Videos

Use the prompt bar at the bottom of your canvas to create AI-generated videos.

Using the prompt bar​

Step 1: Open the Video tab​

The prompt bar sits at the bottom center of your canvas. Click the Video tab.

Step 2: Choose a model and duration​

Use the model selector to pick a video model. Each model offers different durations, aspect ratios, and capabilities. Select the duration that fits your scene.

Step 3: Set the aspect ratio​

Choose an aspect ratio for your video. Available ratios depend on the selected model and include: 1:1, 3:4, 4:3, 9:16, 16:9, and 21:9.

Step 4: Add source inputs​

Depending on the model, you can provide:

  • Start frame — An image to use as the first frame (image-to-video)
  • End frame — A target last frame (supported by some models)
  • Video source — An existing video to edit or transform (video-to-video)
  • Image and video references — Additional assets for style guidance

Step 5: Write your prompt​

Describe what you want the video to show. Each model has a character limit displayed in the input area.

Step 6: Generate​

Click the generate button. The credit cost is displayed on the button before you submit.

Generation types​

TypeDescriptionSource input
Image to videoAnimate a still image into a videoStart frame, optional end frame
Video to videoEdit or transform an existing videoVideo source
Text to videoGenerate a video from a text prompt aloneNo source needed

Not all models support all generation types.

Multi-shot prompts​

Some models support multi-shot generation, where you write separate prompts for different segments of the video:

  1. Enable multi-shot mode in the generation input
  2. Add individual shots, each with its own prompt and duration
  3. The total video duration is the sum of all shot durations

Multi-shot is useful for videos with distinct movements or transitions within a single generation.

Audio generation​

Some models support audio generation. When available, you can toggle audio on or off. The AI generates a matching audio track alongside the video.

Models with custom voice support let you choose from a list of available voices.

Prompt modes​

Like image generation, the prompt bar supports two modes:

  • Fine-tune (default) — Your prompt is enhanced for better results
  • Raw — Your prompt is sent exactly as written

Toggle between modes using the prompt mode selector.

Cost display​

The generate button shows the credit cost before you submit:

  • Video cost = model cost per second × duration
  • Some capabilities (references, multi-shot) add incremental costs
  • If you don't have enough credits, the button turns red and links to the billing page

Model capabilities​

Each video model has different capabilities. The prompt bar adapts its options based on the selected model:

CapabilityDescription
Start/end frameProvide images to guide the first or last frame
Image referencesAdd reference images for style guidance
Video referencesAdd reference videos
Multi-shotWrite separate prompts per shot
Audio generationGenerate a matching audio track
Custom voiceChoose from available voice options
Video editingEdit or transform existing videos

When a model doesn't support a capability, that option is hidden from the interface.

Tips for better results​

Write clear prompts​

Be specific about what you want:

  • Describe the action, mood, and visual style
  • Include details about camera movement if relevant
  • Mention the pace or rhythm you want

Provide good source frames​

For image-to-video, the quality of your start frame matters:

  • Use high-quality images that represent the first moment of your shot
  • Include an end frame to guide where the video should go

Try different models​

Each model produces different styles and qualities. Experiment with several to find what works best for your scene.

Use multi-shot for complex scenes​

When your scene has multiple beats or camera movements, multi-shot gives you finer control over each segment.

Troubleshooting​

Generation takes too long​

Video generation can take a few minutes. If a model shows high demand, consider trying a different one.

Video doesn't match expectations​

  • Add more specific details to your prompt
  • Provide better source frames
  • Try a different model
  • Switch to Raw mode if prompt enhancement is altering your intent
  • Use multi-shot to break down complex movements

Prompt bar not visible​

The prompt bar is at the bottom center of the canvas. Make sure you're in Pre-Production (Canvas) mode.

Learn more​