Kling 4.0 Flash
Kuaishou's cinematic video model, producing silent 720p clips from a text description in about a minute, reached without your name or IP.
Updated November 9, 2026
A humpback whale gliding through golden clouds above a 1950s seaside town at dusk
At a glance
| Maker | Kuaishou |
|---|---|
| Strengths in the app | Cinematic |
| Arena Elo (Artificial Analysis) | 950 |
| Typical wait | about 66 s |
| Who can use it | Paid plans |
| How it’s paid | Credits, from the first use |
| Privacy | Anonymized: NeverNo sends the request to its maker without your name, account or IP address. The maker still reads the text, applies its own rules and may keep it for a while under its own policy. |
Price
What the app charges, read from the same table the servers bill with. 1 credit = $0.01.
| 5-second clip | 118 credits ($1.18) |
|---|---|
| 10-second clip | 235 credits ($2.35) |
What it is
Kling 4.0 Flash is a video model from Kuaishou, offered in NeverNo's Video Studio on paid plans and paid in credits. It makes clips of 5 or 10 seconds in 16:9 or vertical 9:16, at 1280 by 720 pixels, from a text description. The Studio gives it the cinematic tag, and it is a model chosen for its camera work: movement, depth and lighting that feel like a shot rather than a still brought to life.
Its profile is simple. It makes no soundtrack, so it is not tagged audio, and it has no starting-image option or ending image in the Studio. A typical 5 second clip takes about 66 seconds, measured on Runware, our video provider, and steady from one run to the next. That sits between Veo 3.1 Fast (about 36 seconds) and Wan 3.0 (about 144 seconds), and well above the 16 seconds of P-Video 2 Pro or MiniMax H3 Max.
We are careful about its rating. Kling 4.0 Flash was not in the first 25 places of the Artificial Analysis video arena when we looked, so the Studio does not quote an Elo for it and uses a cautious value of its own when ranking models in Auto. Compare it by eye instead, by running the same prompt through Kling 4.0 Flash, Wan 3.0 and Seedance 2.5, which are the other cinematic models in the catalog.
It is anonymized. NeverNo sends the request to Kuaishou without your name, account or IP address; Kuaishou still reads the prompt and applies its own rules, and requests it does not accept are declined. The finished clip is stored in your browser, not on our servers. Because it is silent and text-driven, it rewards a photographer’s vocabulary. Talk about focal length, depth of field, golden hour, backlight and slow shutter feel, and keep the subject simple so the camera has room to act. A sequence of three clips from the same prompt family, one wide, one medium and one detail, cuts together well into a short montage once you add music in your editor. Use the ten second length when the movement is slow and graceful, and five seconds when you want a brisk reveal. Architects, real-estate agents and landscape photographers can test how a space or a view might feel in motion, from a slow reveal of a facade to a drifting shot above a quiet lake. Keep a small notebook of lens, light and motion phrases that produced pleasing frames, and reuse them like presets across projects. Notice how shadows, reflections and haze behave across the frame, since those details reveal whether the model understood your lighting. Overcast mornings, neon puddles, dusty barns and snowy rooftops make rewarding scenery.
Strengths
Tagged cinematic: a good choice for camera moves, depth and mood.
About 66 seconds for a typical 5 second clip, a figure that stays steady.
720p in both 16:9 and vertical 9:16, in 5 or 10 second clips.
A second cinematic option when you want to compare with Wan 3.0 or Seedance 2.5.
Limits
No soundtrack: the Studio does not tag it audio, so add sound later.
Text description only: there is no starting or ending image in the Studio.
No arena Elo to quote; it was outside the first 25 places when we checked.
Anonymized, not private, and video needs a paid plan, paid in credits.
When to choose it
Landscape and travel shots
A slow aerial glide over a valley at dawn, a coastal road at sunset, a market seen from above. Cinematic motion is where this model earns its place, and a silent clip suits a music bed you add yourself.
Moody b-roll for video essays
Rain on a window, steam from a cup, a figure crossing a bridge: short atmospheric inserts that you cut over narration, with the sound handled in your editor.
Comparing looks across models
Run one prompt through Kling 4.0 Flash, Wan 3.0 and Seedance 2.5, and keep the version whose lighting and motion suit your project best.
Prompt tips
Lead with the camera: a slow tracking shot, a low angle, a gentle orbit. Cinematic models respond to a clear instruction about the lens and the move.
Add the time of day and the weather in a few words, such as blue hour, light fog, because they set the whole look.
Ask for one subject and one movement. Since there is no sound, describe what the eye should follow.
Frequently asked questions
Does Kling 4.0 Flash make sound?
No. It is not tagged audio in the Studio, so clips come without a soundtrack.
How fast is Kling 4.0 Flash?
About 66 seconds for a typical 5 second clip, measured on Runware, our video provider, and steady from one run to the next.
Can I start from an image?
Not in the Studio. Kling 4.0 Flash works from a text description. Wan 3.0 and Seedance 2.5 take a start and an end image.
Is it private?
It is anonymized: your name, account and IP address are not sent to Kuaishou, but the lab reads the prompt and applies its own rules. Clips are stored in your browser.
Why is there no Elo for it?
It was not among the first 25 places of the Artificial Analysis video arena when we checked, so we do not quote a rating and judge it by comparing outputs.