Skip to main content

​
Overview

Open Canva supports AI video generation alongside image generation. You can create videos from text prompts (text-to-video) or from existing images (image-to-video) using a growing library of video models.

​
How video generation works

1

Select a video model

In the model selector, switch to the Video tab. Pick one or more video models to run.
2

Enter your prompt

Describe the motion and scene you want. Video prompts benefit from action words and camera direction:
  • “A golden retriever running through a field of wildflowers, slow motion, cinematic”
  • “Camera slowly orbiting around a futuristic city at night, neon lights reflecting off wet streets”
3

Choose generation mode

  • Text-to-Video (T2V): Generate a video entirely from your text prompt.
  • Image-to-Video (I2V): Attach a reference image and the model animates it into a video.
Not all models support both modes. The UI indicates which modes each model supports.
4

Generate and review

Click Generate. Video generation typically takes longer than image generation (30 seconds to several minutes depending on the model and duration).Videos appear in your session history with a play button overlay and autoplay on hover.

​
Available video models

Open Canva includes a wide selection of video models:
Model availability updates regularly. Check the model selector in the app for the latest list.

​
Video duration

Video duration limits depend on your subscription plan:

​
Downloading videos

Click the download button on any generated video to choose your format:
  • MP4 — Standard video format, best compatibility
  • GIF — Animated GIF for easy sharing
  • WebP — Animated WebP, smaller file size than GIF
  • PNG — Extract a single frame as a still image

​
Tips for video generation

Describe motion explicitly: Unlike image prompts, video prompts should describe what moves and how. “A bird soaring across a sunset sky” works better than “a bird at sunset.”
Use Image-to-Video for control: If you have a specific starting frame in mind, generate the image first, then use I2V to animate it.
Expect longer generation times: Video models are more compute-intensive. Plan for 30 seconds to a few minutes per generation.

Image Generation

Learn about AI image generation.

AI Models

Browse all available image and video models.

Plans & Pricing

See duration limits and credits by plan.