AI Video

AI Video

AI Video creates cinematic AI video with synchronized audio, from a prompt or from an image. This page walks through the tool and its options step by step.

Step 1: Choose a mode

Open AI Video from the sidebar. Below the prompt, pick how the video starts:

• Text to Video - generate a video from a written description.
• Image to Video - animate a still image you provide.
• Extend Video - continue an existing clip.

Step 2: Pick a model

Open the model selector (it shows your current model, resolution, and duration). Video models are priced per second of output, and higher-quality models require a higher plan:

Step 3: Set format, resolution, and duration

In the same panel choose:

• Format - 16:9 (wide) or 9:16 (vertical).
• Resolution - 720p, 1080p, or 4K.
• Duration - 4s, 6s, or 8s.

Step 4: Add inputs (optional)

For image or extend modes, add an input image (used as the first frame) and up to three reference images, either uploaded or chosen from your Library.

Step 5: Write your prompt and generate

Describe the scene, optionally use Enhance to expand it, then select Generate. Video credits are dynamic: the cost is the model's per-second rate multiplied by the duration, so an 8-second clip costs twice a 4-second clip at the same model. The exact cost is shown on the Generate button before you run, and finished videos appear in your My Videos tab and in My Library.

For how credits and plans work, see Plans, Credits, and Limits.