POST /api/v1/videos/generations with model: kling-4-0. Generate single takes of 3 to 30 seconds from a prompt or a first frame, guide them through up to 8 middle keyframes and an optional last frame. Every video comes with a native audio track.POST /api/v1/videos/generations with model: kling-4-0. Generate single takes of 3 to 30 seconds from a prompt or a first frame, guide them through up to 8 middle keyframes and an optional last frame. Every video comes with a native audio track.
image_url sets the first frame, keyframes adds 1 to 8 middle frames spaced evenly in playback order, and image_end_url ends the take. Frames are not charged.
duration is any integer from 3 to 30 (default 5), at 720p (default) or 1080p. Text-to-video also accepts ratio 16:9, 9:16, 1:1 or 21:9.
sound and generate_audio are rejected. Describe sound and what to avoid in prompt (up to 2500 characters).
kling-4-0 and a prompt of up to 2500 characters describing scene, motion and camera.image_url. 1 to 8 middle frames; all frames together at most 10 images and at most duration + 1.720p (default) or 1080p; quality is an alias. ratio 16:9 (default), 9:16, 1:1 or 21:9, text-to-video only.sound, generate_audio, negative_prompt, model_params.multi_shot, image_urls and 4K are rejected before any tokens are charged.duration and resolution matter.cost field in the response is what was deducted; failed jobs are refunded automatically.Authorization: Bearer <key> from your server.duration + 1.{"model": "kling-4-0", "prompt": "…", "image_url": "https://…", "duration": 10, "resolution": "1080p"} to /api/v1/videos/generations.img_uuid and poll GET /api/v1/jobs/detail or pass custom_callback_url, then download the video with its audio track.POST https://api.goenhance.ai/api/v1/videos/generations with model: kling-4-0.
Up to 8 middle frames, plus first and last frames, with at most 10 images in total and at most duration + 1.
No. Every video includes a native audio track, and sound / generate_audio are rejected.
Not on this model; image_urls is rejected. Use kling-4-0-refs for reference-to-video.
10.205 tokens ($0.2041) per second at 720p and 15.275 tokens ($0.3055) per second at 1080p, for both text-to-video and image-to-video.
No. Resolution is 720p or 1080p; 4K is rejected before any tokens are charged.