Create coherent 10-second video with text and image direction, then continue in the full Kling 3.0 Omni workspace for frames, Elements, feature video, and storyboards.
A unified AI video model built for creators who want to direct more than a single shot from a single prompt.
Kling 3.0 Omni brings text, images, reusable Elements, motion references, frames, storyboards, and audio direction into one connected workflow. Instead of forcing every visual decision into a long prompt, you can assign the right kind of reference to each creative choice. This keeps complex ideas readable while preserving practical creative control from shot to shot.
Use the quick start above for a simple idea, then continue in Create when a project needs detailed timing or continuity. The free preset produces one 10-second, 720P, silent video per account each day, with no watermark.
Both models create modern AI video. The practical difference is how much reference control and multi-shot direction your project needs.
| Capability | Kling 3.0 Omni | Kling 3.0 |
|---|---|---|
| Primary workflow | Both models create modern AI video. The practical difference is how much reference control and multi-shot direction your project needs. Unified text, image, video, Element, and storyboard direction | Direct prompt-led text-to-video and image-to-video creation |
| Reference control | Broader multimodal control for identity, motion, frames, and scenes | A simpler path for prompts and standard visual references |
| Character consistency | Elements 3.0 supports reusable people, subjects, and objects | Best suited to shots with fewer recurring reference demands |
| Sequence planning | Storyboard controls can organize several timed shots | Efficient for a focused shot or straightforward sequence |
| Best fit | Campaigns, character stories, product films, and directed edits | Fast ideation, social clips, mood shots, and simple animation |
Give each input a clear role, from visual identity and composition to movement and the final frame.
A concise prompt can define intent, action, camera language, pacing, and mood. Reference images anchor appearance; Elements preserve recurring identity; start and end frames guide composition; and a feature video communicates motion that is difficult to describe in words.
Describe subject, action, environment, camera, timing, and desired ending.
Ground the scene with a person, product, setting, style, or composition.
Set the opening image, closing image, or both ends of a transition.
Keep named characters and objects recognizable across separate shots.
Use a short clip when movement and rhythm matter more than a still image.
Build reusable character or object references from multiple views, then call them by name inside a prompt or storyboard.
Each Element combines a short description with two to four clear images. A front view should establish the subject, while the remaining views add useful information about face, clothing, shape, materials, logos, or moving parts.
Up to three Elements can support one task. Keep each source set consistent in age, wardrobe, lighting, and product variant so the model is not asked to reconcile conflicting identities.
Use complementary angles rather than repeated versions of the same image.
Give every recurring person or object a short, unambiguous identity.
Direct interactions among several consistent subjects in one sequence.
Reference the same Element in prompts and custom storyboard shots.
Preserve the details that must survive changes in action and camera angle.
Choose still frames for visual endpoints or a feature video when the essential reference is movement.
A start frame establishes the first composition. An end frame defines where the scene should arrive, making the pair useful for reveals, transformations, and planned camera moves. A feature video instead transfers motion cues such as gesture, tempo, or choreography.
Lock the opening subject placement, styling, and camera position.
Define the closing state for a controlled visual transition.
Reference movement, performance, rhythm, or shot behavior from a clip.
Plan a longer idea as a sequence of timed shots instead of relying on one uninterrupted paragraph.
Custom storyboards divide the request into clear beats. Each shot can define duration, action, framing, camera movement, and recurring Elements, while the complete sequence keeps a shared visual direction.
Allocate the available duration to the moments that carry the story.
Specify action and camera language separately for each beat.
Reuse Elements and visual intent across cuts in the same sequence.
Shape the final delivery with duration, aspect ratio, resolution, audio settings, and watermark controls.
Kling 3.0 Omni supports clips from 3 to 15 seconds, portrait or landscape framing, 720P and 1080P output, and optional native audio. Available combinations depend on the selected workflow; the current free preset remains fixed at 10 seconds, 720P, silent, and watermark-free.
Match clip length to a single action, transition, or compact sequence.
Choose landscape, portrait, or square formats for the intended channel.
Choose 720P or 1080P delivery to match preview, publishing, and production needs.
Enable supported sound output for projects that need synchronized atmosphere.
Use the richer reference workflow where identity, movement, or sequence continuity matters to the result.
Keep a recurring subject recognizable while scenes, actions, and camera angles change.
Anchor shape, branding, materials, and key views across several advertising shots.
Plan short vertical scenes with clear beats, strong openings, and deliberate endings.
Translate a performance or camera reference into a new visual setting.
Test framing, transitions, and sequence structure before committing to production.
Combine visual anchors with precise prompt direction for stylized short-form video.
Move from a quick idea to a finished task in one connected workflow. Start on this model page, add only the controls your shot needs, then continue in Create for full production and review.
Use a suggested prompt or describe the subject, action, camera, mood, and ending in the generator above.
Open the generatorKeep the first draft focused. Name the subject, movement, setting, camera behavior, and the final visual state.
Read the prompt guideAdd an image, Element, start or end frame, feature video, or storyboard only where visual evidence improves control.
Send the draft to the unified Create workspace to set roles, timing, aspect ratio, Elements, and multi-shot direction.
Open the Create workspaceRun the task, inspect continuity and timing in My Creation, then change one prompt instruction or reference at a time.
View My CreationQuick answers about free access, references, duration, and the most useful workflow choices.
Quick answers about free access, references, duration, and the most useful workflow choices. It is a multimodal AI video model that combines prompt direction with images, Elements, frames, feature video, storyboards, and supported audio controls.
Yes. After Google sign-in, each account can generate one free 10-second, 720P, silent, watermark-free video per day.
The workspace supports prompts, general reference images, role-based start or end frames, reusable Elements, and feature video in compatible workflows.
Use frames to control the beginning or ending composition. Use Elements when a person or object must remain recognizable across changing shots.
Supported tasks can use durations from 3 to 15 seconds. The free daily preset is fixed at 10 seconds.