Reference-led scenes
Use images to guide characters, products, outfits, locations, or style.
A multimodal Kling model for text, image, and video-guided generation where references, edits, and continuity need to work together.
Use Kling Video O1 when a prompt alone is not enough and the shot needs reference images or video context.
Use images to guide characters, products, outfits, locations, or style.
Video references can help communicate motion, camera behavior, or an existing scene.
Describe changes conversationally instead of rebuilding every detail from scratch.
O1 performs best when each reference has a clear job and the prompt states the desired change.
Start with the reference image or video that carries the most important identity or motion cue.
Say what should change, what should stay stable, and how the camera should move.
Check subject identity, lighting, and whether the action still reads after the edit.
Use these notes as production guidance inside Creativ Studio. Provider availability and exact limits can vary by account, region, and model rollout.
Reference-to-video and video editing workflows
Text, images, and video references
Creativ Studio exposes 3-15 second options
720p and 1080p through Standard and Pro modes
Keep reference roles explicit to avoid mixed visual intent
Use these as starting shapes, then swap in your own subject, references, and constraints.
Use the reference character and outfit. Move the scene to a rainy alley at night, keep the face recognizable, slow handheld push-in.
Keep the same camera path from the reference video, replace the background with a quiet gallery, preserve subject timing and posture.
It is built for multimodal video work, using text, images, and video context in the same creative pass.
Yes. Describe the edit and clearly state which parts of the reference should remain stable.
Use only the references that serve the shot. Too many unrelated references can weaken the direction.