Skip to main content

AI Video Generation

Generate videos directly from text prompts, with no reference image required.

How to Use​

  1. Select a video model from the dropdown
  2. Type a prompt describing the video you want
  3. Choose an aspect ratio and duration
  4. (Optional) Select a resolution if the model supports it
  5. Set the count (1-4 videos)
  6. Click Generate

Example Prompts​

  • "A drone shot flying over a futuristic city at sunset, neon lights reflecting in the rain"
  • "A girl dancing on a beach, golden hour, slow motion"
  • "A product showcase of a perfume bottle rotating slowly on a marble surface"
  • "Time-lapse of a flower blooming, macro lens, soft focus background"

Available Models​

Text-to-video models include LTX 2.5, MiniMax H3, Happy Horse, Wan3.0, Vidu Q3, and various OpenRouter models (Seedance, Kling, Hailuo, VEO3, Sora 2 Pro).

Some models are available for both image-to-video and text-to-video — they appear in both tools.

Video Settings​

  • Duration — Ranges from 5s to 15s depending on the model
  • Aspect Ratio — Portrait (9:16), Landscape (16:9), Square (1:1), and more
  • Resolution — Available on select models

Computing Power​

Video generation costs more computing power than images. The cost scales with duration and resolution. Check the CP cost in the model dropdown before generating.