Text → Video
Select it with"mode": "text_to_video", or leave mode out and the inputs decide.
Required: promptOptional:
aspect_ratio, duration_seconds, generate_audio, shot_type, multi_prompt
Resolution tiers: standard, pro, 4k (send as tier).
Image → Video
Select it with"mode": "image_to_video", or leave mode out and the inputs decide.
Required: prompt, first_frameOptional:
last_frame, duration_seconds, generate_audio, shot_type, multi_prompt
Resolution tiers: standard, pro, 4k (send as tier).
Reference → Video
Select it with"mode": "reference_to_video", or leave mode out and the inputs decide.
Required: promptOptional:
first_frame, last_frame, reference_images, elements, aspect_ratio, duration_seconds, generate_audio, shot_type, multi_prompt
Resolution tiers: standard, pro, 4k (send as tier).
Restyle from Video
Select it with"mode": "video_reference"; mode is required, because without it these inputs select Edit Video (video_edit).
Required: prompt, source_videoOptional:
reference_images, elements, aspect_ratio, duration_seconds, keep_audio
Resolution tiers: standard, pro (send as tier).
Edit Video
Select it with"mode": "video_edit", or leave mode out and the inputs decide.
Required: prompt, source_videoOptional:
reference_images, elements, keep_audio
Resolution tiers: standard, pro (send as tier).
Reference inputs take asset ids from
POST /v1/uploads or an earlier output’s asset_id.
Prices and allowed values come from the live catalog; GET /v1/models/kling-o3-fal is always current.