LLumoris/ api
v1
POST/v1/generate/video

Text to video

Submit a text-to-video generation job (Veo 3.1 Fast / Lite, Kling 3). Audio is always on.

asynct2v
Parameters
modelstringrequired

veo-3.1-fast | veo-3.1-lite | kling-3

promptstringrequired

Scene description

durationint

Veo: 4 | 6 | 8 · Kling: 3–15 (default 8)

generateAudiobool

Always true — audio included

width / heightint

Snap to 720p/1080p presets (aspects vary by model)

resolutionstring

720p | 1080p (Kling also maps into modelSpecificPayload)

negativePromptstring

Optional — what to avoid

seedint

Optional reproducibility

  • Veo aspects: 16:9 / 9:16 · Kling adds 1:1 (all in 720p / 1080p).
  • generateAudio is forced true — output is video/mp4 with audio.
  • Responses include videoMetadata (fps, width, height, totalFrames).
curl https://api-studio.lumorislabs.com/v1/generate/video \
  -X POST \
  -H "Authorization: Bearer sk-lumoris-..." \
  -H "Content-Type: application/json" \
  -d '{"model":"veo-3.1-fast","prompt":"blonde russian in bar with the monster can cinematic commercial fast movement shots from different angles","duration":8,"generateAudio":true,"width":1280,"height":720}'
{ "jobId": "83253cd6...", "status": "queued" }