Skip to main content
Generate or edit one video per task with synchronized audio. Start from text, frames or reference media, or extend a source video. Request 3–15 seconds of video, or 3–10 seconds for paired frames and extension. Edits follow the source clip, up to 10 seconds on GENGEN. Use HTTPS URLs or authorized GCS objects for media inputs. Retrieve the task to refresh output links during the seven-day retention period.

Model ID

Pass this exact public ID in the top-level model field.

Supported modes

  • text_to_video
  • video_modify
  • image_first_frame
  • image_first_last_frame
  • multimodal_reference
  • video_extend

Inputs

  • Text
  • Image
  • Video

Outputs

  • Video

Capabilities

  • Not specified

Additional capabilities

  • Video generation
  • Video editing
  • Native audio

Technical details

Availability and final customer prices can change. Check the Dashboard model catalog before production use.
See the Omni API guide for examples, supported modes, media limits and billing. Retrieve results with the shared task endpoint. See the API reference overview for authentication, request conventions, and response schemas.