
reference-to-video
Idle
Your request will cost $0.42 per run.
For $10 you can run this model approximately 23 times.
One more thing::
Kling Omni Video O1 is Kuaishou's groundbreaking unified multi-modal video model. The Reference-to-Video mode creates new video content based on subject references — maintaining character, prop, and scene identity while generating entirely new creative scenarios.
Build subjects from multiple reference viewpoints:
Advanced feature extraction ensures:
Generate entirely new content while preserving identity:
| Duration (s) | video = "" | video ≠ "" |
|---|---|---|
| 5 | 0.42 | 0.63 |
| 10 | 0.84 | 1.26 |
Upload Reference Images Provide one or more images of your subject (character, object, or scene).
Describe the Scenario Write a prompt for the new video content.
Example: "The character walking through a futuristic city at night, neon lights reflecting on wet streets"
Set Parameters Choose duration, resolution, and output format.
Generate Receive video with your subject in the new scenario.
kwaivgi/kling-video-o1-std — Video Edit — Edit videos with natural-language instructions for precise, context-aware changes like object removal, scene adjustments, and style refinement while preserving motion consistency.
kwaivgi/kling-video-o1-std — Image to Video — Animate a single image into a high-quality video clip with smooth motion and coherent scene continuity, perfect for marketing creatives and social content.
kwaivgi/kling-video-o1-std — Text to Video — Create videos directly from text prompts with strong prompt adherence and cinematic motion, great for rapid prototyping, ads, and creative concept exploration.