Kling 3.0 · Text and image to video
Turn text and images into videos with Kling 3.0 Standard or Pro. Choose a 5- or 10-second clip, generate, and preview your video.
The video follows your source image’s aspect ratio.
Sign in to create your video
Your saved tasks will appear here.
Describe the motion you want, the camera movement, and the details that should stay unchanged.
View prompt examples →Made with Kling 3.0
See what a prompt can become. Explore four videos created with Kling 3.0 Pro, then try your own scene.
Nature & travel · Pro · 5s · 16:9 · Without sound
A sailboat, soft morning mist, and a gentle camera push-in.
Cinematic wide shot of a small ivory sailboat gliding across a calm alpine lake at sunrise. Gentle ripples trail behind the boat, wisps of mist float above the water, distant mountains catch soft golden light. Slow smooth camera push-in, realistic water reflections, natural colors, one continuous shot, no text or logos.
Product & brands · Pro · 5s · 16:9 · Without sound
Warm studio light and a slow rotation show an amber glass bottle.
Premium product film of an unbranded amber glass perfume bottle on a cream stone pedestal. The bottle rotates slowly on its vertical axis while warm side lighting creates elegant reflections in the glass. Soft beige studio background, subtle shadow, fixed close-up camera, realistic materials, stable bottle proportions, one continuous shot, no letters or logos.
Cinematic storytelling · Pro · 5s · 16:9 · Without sound
A smooth tracking shot follows a figure through a reflective night street.
Cinematic night street in a modern city after light rain. A lone person in a dark coat walks away from the camera along the sidewalk. Warm café windows and cyan lights reflect in wet pavement, subtle steam drifts from a street vent. Smooth slow tracking camera follows behind, realistic movement, rich contrast, one continuous shot, no readable signs or logos.
Lifestyle & social · Pro · 5s · 16:9 · Without sound
A side-follow camera captures a cyclist moving along the coast.
Lifestyle film of an adult cyclist in a mustard yellow jacket riding a simple bicycle along a seaside path at golden hour. Smooth side-follow camera keeps the cyclist centered as sunlit grasses move in the breeze and the ocean glitters in the background. Natural pedaling motion, realistic anatomy and bicycle geometry, warm cinematic colors, one continuous shot, no text or logos.
Kling 3.0 credit usage
Credits = video duration in seconds × the rate for your model and sound setting. Standard and Pro estimates are shown in the generator. Sound, 4K, Omni, and Motion rates are listed for reference; these options are not enabled in the current workspace.
The 40-credit signup offer covers one shot at this rate. A 5-second Pro shot without sound uses 50 credits.
| Model | Sound | Credits / second | 5-second estimate |
|---|---|---|---|
| Kling 3.0 Standard | Without sound | 8 | 40 credits |
| Kling 3.0 Standard | With sound | 12 | 60 credits |
| Kling 3.0 Pro | Without sound | 10 | 50 credits |
| Kling 3.0 Pro | With sound | 15 | 75 credits |
| Kling 3.0 4K | Not specified | 40 | 200 credits |
| Kling 3.0 Omni | Without sound | 8 | 40 credits |
| Kling 3.0 Omni | With sound | 10 | 50 credits |
| Kling 3.0 Pro Omni | Without sound | 10 | 50 credits |
| Kling 3.0 Pro Omni | With sound | 12 | 60 credits |
| Kling 3.0 Standard Motion | Not specified | 12 | 60 credits |
| Kling 3.0 Pro Motion | Not specified | 15 | 75 credits |
A price listing does not confirm model availability. Kling 4.0 pricing will be announced separately. View credit packs or read the usage guide.
Understand the model and the workspace
Kling 4.0 provides access to the published Kling 3.0 model family. Kuaishou's official materials describe native audiovisual output, stronger subject consistency, and storyboard controls. On this site, create 5- or 10-second videos from text or one image using Standard or Pro.
The official Kling 3.0 guide covers text-to-video and image-to-video. In this editor, choose words to define a new scene or select one still to anchor its appearance. Your motion description remains important in either case: a source image describes a starting look, while the prompt explains the action.
The official model family supports native audiovisual generation and clips up to 15 seconds. Those are model-level capabilities. This generator supports 5- and 10-second clips without sound. Plan one visual moment and add music or dialogue in your editing workflow.
Kuaishou describes consistency and multi-shot storytelling as part of the 3.0 family, with more detailed storyboard control in Video 3.0 Omni. The current editor here does not expose Omni, element libraries, or multi-shot controls. Keep your prompt to one visible action and decide which subject details must remain stable.
The fields you can prepare here
Each choice should solve a composition problem. Select the source that describes the subject best, then choose a frame that gives the action enough room. The table shows the current generator, rather than the full specification of every Kling model or API.
| Control | Current choice | Practical guidance |
|---|---|---|
| Starting material | Text or one JPG, PNG, or WebP image | Use words for a new scene; use a still when appearance or composition matters |
| Prompt | A written shot description | Describe the subject, action, camera, and details that should stay stable |
| Wide frame | 16:9 | Useful for a landscape, a product with surrounding space, or movement across a scene |
| Vertical frame | 9:16 | Useful for a standing subject or a composition intended for a phone screen |
| Duration | 5 or 10 seconds | One gesture or reveal usually needs less explanation than several scene changes |
| Preview | Returned video with playback controls | Review the completed clip and open the video file to save it |
Framing changes what can remain visible. If a cyclist moves across a wide scene, 16:9 may give that action more room. If the subject is a standing person, 9:16 may fit the silhouette better. Choose the ratio before polishing the prompt so your instruction matches the available space.
A practical four-step workflow
Plan the shot in the same order you would explain it to a collaborator. Choose your input, model, and duration, then generate and review the video.
Start with text if the scene is new. Select a permitted still if you want a recognizable subject or composition. Check the crop, lighting, and visible details first; a cluttered image makes it harder to identify the part that should move.
Write one main action with a concrete verb: turns, walks, lifts, or rotates. Give the action a clear ending. Avoid a list of unrelated events, and mention the features that matter most if the prompt starts from an image.
In text mode, choose 16:9, 9:16, or 1:1, then describe the viewpoint. In image mode, prepare the source image’s crop. Keep the camera fixed if the action already tells the story, or add one purposeful move. Separate subject motion from camera motion so both instructions can be understood.
Check the credit estimate and select Generate. Sign in if prompted. Follow the task in Output Preview, play the completed clip, and open the video file to save it.
Concrete starting points
These suggested prompts demonstrate how to describe one action, a viewpoint, and a stable detail. They are writing examples, with no claimed generated result. Replace the subject and setting with your own material, then review whether the action fits five seconds.
Suggested prompt · Input example
A person in a red jacket stands beside a city café at dusk and turns toward a passing bicycle. Fixed eye-level camera in a vertical composition. Keep the person fully visible and the café window lights steady as the bicycle passes.
The gesture gives the scene a clear event without requiring a second location. The fixed camera keeps attention on the reaction. In a vertical frame, leave enough space around the person for the turn and any text you plan to add later.
Suggested prompt · Input example
A plain ceramic bottle stands on a stone plinth against a soft gray background. The bottle rotates slowly to show its side profile while the camera stays fixed. Gentle side lighting reveals the surface texture. Keep the bottle's proportions and the plinth unchanged.
Start with your own product image if appearance matters. The prompt separates object movement from camera behavior and names the stable geometry. Review the silhouette and surface details in any later output, and add exact branding through a controlled editing step.
Suggested prompt · Input example
A traveler waits on a quiet seaside railway platform at sunrise. A light breeze moves the coat as the traveler looks toward the tracks. Wide composition with a slow camera push-in. Keep the horizon level and the platform layout consistent throughout the shot.
This prompt uses a small subject action and one camera movement. The setting establishes the mood without a scene change. Make sure the proposed push-in does not crop out the detail that gives the shot its meaning, such as the tracks or coastline.
Build a prompt around the deliverable
A short scene can support a larger creative project. Decide where the clip would be used before writing the action, and keep the later editing and review steps in mind. These are suggested workflows for planning, not performance guarantees.
Plan a moment that works without a long explanation: a reaction, an arrival, or a visual reveal. Keep the focal subject away from the edges if the final platform adds interface overlays. Prepare alternative prompts that change one creative element at a time, such as the lighting or camera distance.
Use one image to anchor a product's appearance, then describe a restrained movement or lighting change. A presentation concept needs a clear view of the object, not a crowded background. Review exact labels, materials, and proportions separately before considering a clip suitable for a brand deliverable.
Turn a script beat into a shot description a team can discuss. Name the camera height, subject action, and ending position. If a sequence needs several shots, keep them as separate plans; generate each shot separately, then assemble your sequence in a video editor.
Choose the page that answers your question
Kling 4.0 has been officially released, but this site has not connected it. Use Kling 3.0 here for text or single-image generation; read the 4.0 page for the official model specifications and current site availability.
| Your task | Start here | What to prepare |
|---|---|---|
| Write a specific shot | Kling 3.0 | Subject, action, camera, ratio, and a five-second plan |
| Animate a defined appearance | Kling 3.0 | One permitted still and a precise motion instruction |
| Read the released 4.0 specifications | Kling 4.0 information page | Official capabilities, Flash limits, and features still upcoming |
| Learn the basic workflow | How to Use | A simple idea you can turn into one continuous shot |
Keep the creative idea reusable. A good prompt states what the viewer sees and what changes over time; it does not depend on an unconfirmed resolution, model tier, or price. Save those production settings as a separate checklist until the actual workflow is available.
Inputs, settings, and results
Choose Text to Video or Image to Video, select Standard or Pro, and add your inputs. Sign in and select Generate. Your task and returned video appear in Output Preview.
Use text when you want to describe a new scene. Use one image when the appearance or composition is already defined. In both cases, write the intended motion and camera behavior; the image does not replace the action instruction.
Write enough to specify the subject, one action, camera view, and important stable details. A short clear instruction is easier to review than a long list of moods, effects, and conflicting movements. Split unrelated actions into separate shot plans.
Official model documentation describes capabilities beyond this site's editor, including audiovisual generation. This generator supports 5- and 10-second videos without sound. Read the linked official guide for model information, and use the visible fields for this site's available choices.
The current generator accepts one reference image. It does not expose multiple references, an end frame, video references, or an element library. Choose a still that shows the essential subject and describe the motion you want.
Watch the entire clip for recognizable subject details, natural motion, stable geometry, and an ending that fits the prompt. Check labels and brand elements separately. An attractive first frame does not establish that the whole shot is usable.
Put your next shot into words
Choose one subject and one action. Add the camera view, decide between a wide and vertical composition, and check that the idea fits five seconds. The suggested prompts above can help you get past a blank editor.
Add your text prompt or source image, choose Standard or Pro, and generate your next video. Review the complete clip before adding it to your project.
