Kling Text-to-Video
Kling O1 text-to-video turns prompts into cinematic 5–10s clips with accurate physics and strong prompt adherence. 60 credits per second in 16:9, 9:16 or 1:1.
Kling Text-to-Video is the prompt-only mode of Kling Omni Video O1, Kuaishou's unified multimodal video model. It turns natural-language descriptions into cinematic 5 or 10-second clips with accurate physics, strong semantic understanding and a film-like look. On Conut.ai it is billed at 60 credits per second and outputs 16:9, 9:16 or 1:1, with no image upload required.
- Duration: 5s / 10s
- Aspect ratio: 16:9 / 9:16 / 1:1
- Price: 60 credits/s
- Input: Prompt only
How to use Kling Text-to-Video
Write for physics and camera language. Kling O1's strength is grounded motion, not abstract mood boards.
- Pick the frame shape Choose 16:9 for cinematic and YouTube, 9:16 for shorts, or 1:1 for feed posts. Changing ratio after generating means regenerating.
- Write the action in order Start with the subject, then the action, then the camera move. Keep it to a few seconds of screen time rather than a whole scene.
- Add physical detail Name what moves and how: fabric lifting, water splashing, dust settling. Kling responds to real-world motion cues better than style adjectives.
- Generate and inspect motion A 5-second test costs 300 credits. Watch hands, feet and contact points, where physics errors are easiest to spot.
- Extend only after it holds If the 5-second version works, rerun at 10 seconds with the same prompt. Do not add new story beats to the longer take.
Settings reference
Kling O1 Text-to-Video is prompt-only: duration and aspect ratio are the only generation controls, and billing is linear per second.
| Setting | Value | Notes |
|---|---|---|
| Prompt | Required | Describe the scene, action and camera movement in natural language. |
| Duration | 5s / 10s | Default 5s. 5s = 300 credits, 10s = 600. |
| Aspect ratio | 16:9 / 9:16 / 1:1 | Default 16:9. Pick for the destination before writing the prompt. |
| Price | 60 credits/s | Billing scales linearly with duration. |
| Audio | None | Silent video. No sound toggle on this endpoint. |
| Resolution | Model default | No resolution selector, seed or negative prompt. |
What Kling O1 text-to-video is good at
Kling O1 is built for shots that need to look like film rather than a generated demo. Its motion has weight, its camera moves follow the prompt, and its lighting and color grade lean cinematic by default. That makes it a practical choice for trailers, fashion films, music-video inserts and mood pieces.
The model's semantic understanding is the other strength. It handles compound prompts with several subjects, spatial relationships and a specific camera behavior better than older text-to-video models, so you spend less time restarting and more time refining a shot that already works.
- Film-grade visual style out of the box
- Accurate real-world physics and weight
- Strong comprehension of detailed prompts
- Three aspect ratios for horizontal, vertical and square
- 5 or 10-second clip lengths
Kling text-to-video pricing
Billing is per second, so the cost scales linearly with clip length. A 5-second clip costs 300 credits and a 10-second clip costs 600. There is no resolution selector on this endpoint, which keeps the price predictable: you are choosing duration and aspect ratio, not a resolution tier.
- 60 credits per second
- No resolution picker on this endpoint
- Three aspect ratios: 16:9, 9:16 and 1:1
- No audio generation on this text-to-video endpoint
| Duration | Credits | Best for |
|---|---|---|
| 5s | 300 | Prompt and camera tests |
| 10s | 600 | Final social and cinematic shots |
Kling O1 vs O3 Pro vs other text-to-video models
Kling O1 is the value option in the Kling family. O3 Pro costs more per second but adds 3–15 second granular durations and optional sound. Outside the family, Sora 2 brings synchronized audio and long 720p clips, while Seedance 2.5 reaches 4K and 30 seconds.
- Choose O1 when the look matters and 10 seconds is enough.
- Choose O3 Pro when the shot must hit an exact duration or include sound.
- Choose Sora 2 for synchronized audio and long 720p clips.
- Choose Seedance 2.5 for 4K delivery.
| Model | Duration | Price | Audio | Best for |
|---|---|---|---|---|
| Kling O1 T2V | 5s / 10s | 60 /s | No | Cinematic look at a lower rate |
| Kling O3 Pro T2V | 3–15s | 80–100 /s | Optional | Exact timing and sound |
| Sora 2 T2V | 4–20s | 200–1,000 /clip | Synchronized | OpenAI motion and audio |
| Seedance 2.5 T2V | 5–30s | 60–600 /s | Native | 4K and long takes |
| Wan 3.0 T2V | 2–30s | 23–90 /s | Optional | Flexible length on a budget |
Prompting Kling for cinematic results
Kling rewards a clear shot description. Start with the subject and the action, then add the camera, the light and the mood. It understands film language such as close-up, wide shot, dolly in, crane up, rack focus and handheld, and it uses those cues to shape the output.
Physics cues are unusually effective. Words like heavy, weighted, momentum, impact and follow-through help the model produce motion that feels grounded. For character shots, describe what the body is doing rather than only the emotion; posture and gesture give the model something concrete to animate.
- Subject and action first, camera and light second
- Use real shot vocabulary: close-up, wide, dolly, crane, rack focus
- Add one physics cue for weight and impact
- Describe posture and gesture for character shots
- Keep the prompt to one scene and one continuous action
Limitations and when to use another Kling mode
This endpoint does not generate audio, so any clip that needs sound must be paired with a separate audio workflow or generated with Kling O3 Pro, which exposes a sound option. There is also no resolution selector and no seed or negative prompt, which limits fine-grained reproducibility.
If you have a reference image, Kling Image-to-Video preserves it while adding motion. If you need to edit an existing clip, Kling Video Edit is the right mode. If you need sound and exact timing, Kling O3 Pro Text-to-Video is the upgrade.
- No audio on this endpoint
- No resolution selector
- No seed or negative prompt
- 10-second maximum
- For image-led shots, use Kling Image-to-Video
Prompt examples for Kling Text-to-Video
- Cinematic portrait: Close-up of a weathered sailor standing on a dock at dawn, wind moving his coat, camera slowly pushes in as he looks toward the horizon, soft backlight, shallow depth of field, muted teal and amber grade. One expression, one camera move, one light source.
- Action beat: A martial artist spins and delivers a heavy kick into a wooden post, splinters flying, camera orbits the impact at knee height, dust in the air, hard sidelight, high shutter speed. Physics cues such as heavy and impact help the model sell weight.
- Fashion walk: A model walks a rain-slick runway toward the camera, fabric trailing, strobe lights firing behind her, slow push-in, glossy editorial look, cool white and deep black palette. Vertical 9:16 works well for runway and social fashion.
- Fantasy landscape: A lone rider crosses a stone bridge toward a mountain fortress, mist rising from the valley, camera cranes up to reveal the scale of the fortress, overcast light, epic fantasy grade. Use a crane move to reveal scale at the end of the clip.
- Product motion: A glass perfume bottle rotates slowly on a black marble pedestal, liquid catching the light, camera tracks around it at eye level, single hard key light, luxury commercial look. A slow orbit is a safe way to show a product from several angles.
- Noir street: A detective walks through a foggy alley at night, coat collar up, streetlamp pooling light on wet cobblestones, camera tracks behind him, deep shadows, high-contrast noir grade. Tracking from behind avoids face detail while keeping tension.
Pro tips for Kling O1 Text-to-Video
This is the value tier of Kling. Its look is cinematic, but it has no audio and no resolution or seed controls.
- One scene, one shot Write a single camera setup. If you need a cut, generate two clips and edit them; multi-shot prompting belongs to O3 Pro.
- Describe contact and weight Say that a boot presses into gravel or a hand grips a railing. Physical contact cues are where Kling's motion model looks most convincing.
- Avoid text in frame Signage and subtitles are a weak point. Add typography in an editor instead of asking the model to render letters.
- Use 9:16 from the start Vertical clips are common on this tier. Choosing 9:16 up front avoids re-generating for shorts after a successful 16:9 test.
- Plan audio separately Budget for music, voice or ambience in post, or move to O3 Pro when sound must be generated with the picture.
Frequently asked questions
How much does Kling text-to-video cost?
It costs 60 credits per second: 300 credits for a 5-second clip and 600 for a 10-second clip.
How long can a Kling text-to-video clip be?
5 or 10 seconds. Ten seconds is the maximum on this endpoint.
Does Kling text-to-video generate audio?
No. This endpoint produces video only. Kling O3 Pro Text-to-Video exposes an optional sound mode if you need audio.
What aspect ratios are available?
16:9 landscape, 9:16 vertical and 1:1 square.
Can I choose a resolution?
No. This endpoint does not expose a resolution selector; output resolution is set by the model. Pricing depends only on duration.
Does Kling support a negative prompt or seed?
Not on this endpoint. Those controls are not exposed, so results vary between generations even with the same prompt.
What is the difference between Kling O1 and O3 Pro?
O3 Pro is the newer, higher-priced generation with 3–15 second granular durations, stronger physics and optional sound. O1 is the value option for 5–10 second cinematic clips.
Can I animate an image instead?
Yes. Use Kling Image-to-Video, which takes a reference image and adds motion while preserving the subject and composition.
Compare with other text-to-video models
- Kling Image-to-Video
- Kling O3 Pro Text-to-Video
- Kling O3 Pro Image-to-Video
- Sora 2 Text-to-Video
- Seedance 2.5 Text-to-Video
- Browse all AI video models
Specifications
- Provider: Wavespeed
- Model type: Text to video
- Credit pricing: 60 credits per second
- Capabilities: new, popular, multi-shot
About Conut.ai
Conut.ai is an AI image, video and audio generation platform with 50+ creative tools. It supports text-to-image, image-to-video, face swap, motion control, relighting, inpainting and video effects through models including Nano Banana Pro, Seedream, GPT Image, Sora 2, Kling, WAN and Seedance. Start free — no credit card required.