deeprelayDocs
CH·02CLI reference

deeprelay video

Generate and manage videos

Generate and manage videos

Synopsis

Generate a video from a text prompt with a serverless model. Generation runs asynchronously — create returns a queued job immediately (use --wait to poll to completion), get / list track its status and progress, download fetches the finished mp4, and cancel stops an in-flight job. Completed videos are retained for a limited window before they expire.

Billing is per second of output video at the model's catalog rate; the final charge appears as a Cost line (cost_cents in JSON) once the job completes. Failed and cancelled jobs are never billed.

deeprelay video [command]

Examples

# Submit a job and return the queued video immediately
deeprelay video create --model deeprelay/seedance-1.0-lite --prompt "a timelapse of clouds over mountains"

# Block until the video is ready (poll up to 10 minutes), then download it
deeprelay video create --model deeprelay/seedance-1.0-lite --prompt "ocean waves at sunset" --wait 10m
deeprelay video download <video-id> --out waves.mp4

# Track progress and list recent jobs
deeprelay video get <video-id>
deeprelay video list

# Cancel an in-flight job
deeprelay video cancel <video-id>

Options

  -h, --help   help for video

Options inherited from parent commands

      --api-base string   API base URL (override with DEEPRELAY_API_BASE env) (default "https://api.deeprelay.ai/v1")
      --debug             Enable debug logging to stderr
      --no-preflight      Skip the plan/credit check before inference requests (override with DEEPRELAY_NO_PREFLIGHT env)
  -o, --output string     Output format: table|json (default table on TTY, json otherwise) (default "table")

SEE ALSO

← The gpu CLI