Generate video clips from prompts and reference images

What AI video generation supports

BrandReel exposes the inputs and settings supported by its current video-model catalog.

Text to Video. Generate an individual video clip from a written prompt without supplying a frame image.

Reference Frames. Use an uploaded asset or generated image as a start frame, with additional frame support on compatible models.

Model Controls. Select from six video models and the aspect ratios, durations, resolutions, and negative-prompt options each model supports.

Native Audio. Enable model-generated audio on Kling 3.0 and Seedance 2.0 models. Availability and credit cost depend on the selected model.

Each request creates one video clip. BrandReel stores the prompt and render settings with the result so you can inspect, retry, download, or submit the clip to the moderated community gallery.

Choose inputs before generation. Start with a text prompt alone or add compatible frame images from your uploaded assets and earlier image generations.

Settings depend on the model. The full catalog covers landscape, vertical, square, and 4:3 formats with output from 480p to 4K. Every option is available to every account; the credit cost varies by model and settings.

Credits and downloads. Credit cost varies by model, duration, resolution, and audio setting. Completed clips download as MP4 files.

1. 1Write the motion, subject, setting, and camera behavior you want in the prompt.
2. 2Add a reference frame when the selected model supports image input.
3. 3Choose model settings within your plan limits, review the credit cost, and generate the clip.

• Six selectable video models
• Text-to-video and supported frame inputs
• Five aspect ratios across the model catalog
• MP4 downloads for completed clips