od MiniMax
MiniMax H3 Reference-to-Video
Hailuo H3 — reference images, video, and audio for character/scene-consistent 2K video (4–15s)
minimax-h3-ref2vGenerate video with this model.
/api/v1/videos/generationsGET /api/v1/videos/generations{
"model": "minimax-h3-ref2v",
"prompt": "A cinematic drone shot over a coastal city at sunset",
"duration": 4,
"resolution": "1080p",
"aspect_ratio": "adaptive",
"reference_image_urls": [
"https://example.com/character-ref.png"
],
"video_urls": [
"https://example.com/motion-ref.mp4"
],
"audio_urls": [
"https://example.com/voice-ref.mp3"
]
}curl -X POST https://kunya.ai/api/v1/videos/generations \
-H "Authorization: Bearer kunya_YOUR_API_KEY" \
-H "Content-Type: application/json"
-d '{"model":"minimax-h3-ref2v","prompt":"A cinematic drone shot over a coastal city at sunset","duration":4,"resolution":"1080p","aspect_ratio":"adaptive","reference_image_urls":["https://example.com/character-ref.png"],"video_urls":["https://example.com/motion-ref.mp4"],"audio_urls":["https://example.com/voice-ref.mp3"]}'MiniMax
Hailuo H3 — first/last-frame image-to-video at 2K with adaptive aspect ratio (4–15s)
Dokumentacja APIMiniMax
Hailuo H3 Max — fal-optimized fast text-to-video at 480P/768P (5–15s), ~35× faster than standard H3
Dokumentacja APIKunya (Wan)
Alibaba Wan 2.7 — multi-character reference video with voice cloning, 2-10s
Dokumentacja APIKunya (Kling)
Kling V3 — motion transfer from reference video to character in reference image (up to 10s per render)
Dokumentacja API