H3 Max

Create cinematic AI video from text, image, video, and audio references.

H3 Max

H3 Max is a multimodal video workflow for directing cinematic motion from text, images, video, and audio references.

Start a creation

  1. Choose text-to-video, image-to-video, or all-reference mode.
  2. Describe the scene, action, camera movement, lighting, and sound.
  3. Add up to four references when the creation needs visual or motion continuity.
  4. Select resolution, duration, and aspect ratio.
  5. Generate, review, and refine the result.

Available controls

  • Up to 2K output
  • Clips up to 15 seconds
  • Native stereo sound
  • Text, image, video, and audio prompts
  • Multiple aspect ratios for social, cinematic, and commercial work

Use the creator to start a video or review the available plans.