Image to Video (Beta)

The Image to Video Add-on uses AI to generate videos from your images. You provide one or more source images, optionally with a text prompt describing the motion or scene you want. Cloudinary generates a video and saves it to the 'image-to-video' folder in your Media Library. You drive video generation from the Image to Video app in the Cloudinary Console.

Getting access

The Image to Video feature is enabled via an add-on subscription. You can subscribe on a self-service basis from the Add-on Marketplace. If you need help, contact support.

Once your subscription is active, the add-on appears on the Installed Add-ons tab of the Add-ons page in the Console. Depending on your account or product environment setup, you may need to access add-ons from Settings > Add-ons instead.

Plans and quota

Each plan provides a monthly quota of credits, and every video you generate consumes credits from that quota. Enabling Audio consumes twice as many credits as the same generation without audio.

When you run out of credits for the current billing cycle, you can't generate more videos until more credits are available. The Image to Video app displays your remaining credits so you can monitor usage before generating more videos.

For details about prorated billing, mid-cycle upgrades, and downgrade behavior, see Add-on units and billing.

Free trial

You receive a one-time allocation of free trial credits so you can try the Image to Video Add-on before subscribing to an add-on plan. The size of the allocation depends on your Cloudinary plan:

Providing source images

You can provide source images using two modes, selectable via the tabs at the top of the left panel:

• Start & End Frames (default): Add a start frame and optionally an end frame. The generated video transitions from the start frame to the end frame based on your prompt.
• Reference Images: Add a main image and up to two additional reference images to help guide the style of the generated video. Reference Images and End Frame are mutually exclusive, so you'll need to remove an end frame before adding reference images, or vice versa.

To add images, click Add Start Frame, Add End Frame, or Add photo depending on the selected mode.

Cropping a source image

If a source image isn't 16:9 or 9:16, the generated video can include black bars around the image. To avoid them, crop the image to 16:9 or 9:16 in the Image to Video app in the Console. Cropping also lets you focus the video on the part of the image that matters. You can crop a start frame, an end frame, a main image, or a reference image. You crop each image separately, and the app uses any image you don't crop as is. Cropping isn't available for images with a private or authenticated delivery type.

To crop an image:

1. In the left panel of the Image to Video app, hover over the start frame, end frame, main image, or reference image you want to crop, and click the Crop icon to open Crop image.
2. Select 16:9 (the default) or 9:16.
3. Drag the crop rectangle to frame the part of the image you want, and drag a corner to resize it. The rectangle keeps the selected aspect ratio and stays inside the image.
4. Click Save.

You can also use the arrow keys to move the rectangle or a selected corner. To close Crop image without cropping, click Cancel or the X in the top-right corner.

Cloudinary saves the cropped image to your Media Library as a new asset, and the app uses it in place of the original. The original asset doesn't change. The new asset's public ID is the original's with a cropped_ prefix and a unique suffix. For example, cropping products/shoe creates products/cropped_shoe_<unique_id>.

Cloudinary saves the new asset in the root folder of your Media Library, even if the original is in another folder. If your Media Library upload preset sets an asset folder, Cloudinary saves it there instead.

Writing a prompt

Under What do you want to create?, enter a text description of the scene or motion you want to generate. For example, "Turn around" or "add rain".

If you're not sure what to write, open Presets and choose one of the Suggested presets:

• Freestyle Animate
• Zoom In
• Zoom Out
• Pan Left
• Pan Right
• Rotate
• Fade In
• Fade Out
• Slow Motion
• Time Lapse
• Cinematic Movement
• Dramatic Reveal

Selecting a preset populates the prompt field with a description optimized for that type of motion. You can also save and reuse your own prompts under Your Presets via Manage presets.

You can also leave the prompt empty, as long as you've added a start frame (or a main image, in Reference Images mode). The app then creates a subtle, natural animation that keeps your image's subject, composition, and style intact. A prompt is still required when generating videos via the API.

Configuring advanced settings

Expand Advanced Settings to configure the following options:

Generating the video

After configuring your image, prompt, and settings, click Generate Video to start. The right panel displays a loading state while Cloudinary generates the video.

Once the video is ready, the app displays a full preview with playback controls. A notification confirms that the video has been saved to the 'image-to-video' folder in your Media Library.

For example, the following run uses a start frame, an end frame, and the Dramatic Reveal preset:

Start frame End frame

The preset fills in a prompt describing a slow, controlled zoom that builds tension. Clicking Generate Video produces a video that transitions from the start frame to the end frame:

Viewing and sharing creations

The top bar provides the following options:

• Generated Videos: Opens a picker of your previously generated videos. Select one to reopen it and continue editing, as described in Iterating on a generation.
• Share: Shares the generated video.
• Share Feedback: Provides feedback on the Image to Video app.

Iterating on a generation

Each video you generate saves the settings used to create it, so you can refine a result instead of starting over. Open a previous generation from Generated Videos in the top bar, and the app restores that video's source images, prompt, and advanced settings. Adjust any of them and click Generate Video to create a new version.

Generating videos via the API

In addition to the Image to Video app, you can generate videos programmatically using the Image to Video API. You submit an image and a prompt to create a generation job, poll the job until it completes, and receive the generated video's asset details and delivery URL. You can also save and reuse prompts.

The API uses Basic Authentication with your Cloudinary API Key and API Secret, and its base URL is https://api.cloudinary.com/v2/video/{cloudName}.

For the full list of endpoints, request and response payloads, and interactive examples, see the Image to Video API reference.