Kuaishou's groundbreaking unified video creation engine. Generate, edit, and transform videos entirely in a single system — reference-based generation, natural language editing, multi-subject consistency, and skill combos — all without switching tools.
Available on SeedDance platform
Launched on December 1, 2025, Kling O1 is the industry's first unified multimodal video creation tool. Built on the Multimodal Visual Language (MVL) framework, it fuses reference-based generation, text-to-video, start/end frame control, video in-painting, video editing, style re-rendering, and shot extension into one single engine — eliminating the need to switch between separate tools throughout the entire creative lifecycle.
Seven video tasks integrated into one platform: text-to-video, image-to-video, reference-based generation, start/end frame control, video in-painting, style re-rendering, and shot extension. The entire creative lifecycle from inception to refinement is a seamless single-stream workflow.
Kling O1 acts with director-like memory — retaining identity of main characters, props, and settings across all shots. Industrial-grade consistency is maintained through dynamic camera movements, lighting changes, and complex group scenes.
Edit videos conversationally: type 'remove passersby', 'transition day to dusk', or 'swap the protagonist's attire' and Kling O1 executes pixel-level semantic reconstruction — from targeted subject replacement to full-frame style re-rendering — no manual masking or keyframing needed.
Execute compound creative operations simultaneously: insert a subject while modifying the background, generate from a reference image while shifting artistic style. Tasks that previously required multiple tools now happen in a single generation pass.
Kling O1 resolves the most critical pain point in real-world AI video adoption — character and scene inconsistency — while unifying the entire production workflow into a single intelligent system.
Seven integrated video creation capabilities unified in a single AI system — for film, television, social media, advertising, and e-commerce.
Generate cinematic video clips from natural language descriptions. Kling O1's deep semantic understanding produces coherent, visually consistent results for complex narrative prompts.
Upload reference images to lock onto a character's unique visual traits, and generate entirely new scenes with that character consistently preserved. Supports multiple simultaneous reference subjects.
Define both the opening and closing frames of your video — Kling O1 intelligently fills in the motion and narrative between them, giving you precise control over the arc of every clip.
Insert new subjects or remove unwanted elements from existing video with natural language commands. Kling O1 performs pixel-level semantic reconstruction without manual masking.
Edit existing footage conversationally: change backgrounds, swap outfits, alter time of day, modify props — all by describing the change in plain text. No keyframing or manual editing required.
Apply different artistic styles to existing videos — from photorealistic to anime, film noir to watercolor — while maintaining the original motion and composition.
Extend any video clip forward or backward while maintaining complete visual and narrative continuity. Chain multiple clips together to build longer productions from AI-generated segments.
Execute multiple creative tasks simultaneously in a single pass: insert a subject + modify background, or generate from reference + shift style. Complex multi-step workflows compressed into one prompt.
Everything you need to know about Kling O1 and how to use it on SeedDance.
Experience the world's first unified multimodal AI video model on SeedDance. Generate, edit, transform, and extend videos — all in one seamless creative engine.