Reviews
Kling 4.0 Review: Quality, Speed and What a Clip Costs
We ran Kling 4.0 on six prompts at 5 seconds and 720p, with sound: how it handled motion, speech and text, how long it took, and what each clip cost.
By DER TeamPublished Tested
Disclosure: DER is our product. Every model in this post runs on DER, and you pay DER in credits when you use it.

Kling 4.0, from Kuaishou, is the latest Kling model on DER. It turns a text prompt or a photo into a clip of 3 to 30 seconds at 720p or 1080p, with sound generated in the same pass. In our six tests it kept fast body motion steady, spelled the word on the mug correctly and spoke the barista's line word for word; for one continuous shot, say so in the prompt, because Kling 4.0 sometimes cuts to a close-up on its own. A 5-second 720p clip cost 600 credits and came back in about 4 minutes.
What Kling 4.0 does on DER
- Three ways to start: text to video, image to video, and first and last frames.
- Length: any whole number of seconds from 3 to 30.
- Resolution: 720p or 1080p.
- Sound: every clip comes with sound, made together with the picture.
- Price: by the second, from 360 credits for a 3-second clip (≈ $3.60). The Studio shows the exact credits before you start, and a clip that fails costs nothing.
- Access: Kling 4.0 is a Pro model on DER. Tiers come from your total top-ups; there is no subscription.
We ran every clip at 5 seconds, 720p and 16:9, so the numbers below compare like with like. The full setup is in the How we tested box above.
Text to video: five prompts
Each prompt checks one thing video models often find hard: fast body motion, a character speaking on camera, a smooth camera move over a landscape, readable text on an object, and a stylized animated look.
Text to video: five prompts
5 seconds, 720p, 16:9, with sound. Kling 3.0 ran the first three prompts.
Fast human motion: a breakdance spin on a wet street
PromptA young dancer in a yellow jacket performs a fast breakdance spin on a wet city street at night, neon signs reflecting in the puddles, a handheld camera circles the dancer, the slap of sneakers on wet asphalt and a distant hip-hop beat.
Speech and lip sync: a barista talks to the camera
PromptIn a small sunlit café, a barista with curly hair slides a cup across the counter, looks into the camera and says: "Your oat latte is ready. Careful, it is hot." Soft background chatter and the hiss of a milk steamer.
Landscape and camera move: drone over a misty forest
PromptAerial shot gliding forward over a misty pine forest at sunrise, a river winding through the valley below, golden light breaking through the fog, slow and steady drone movement, birdsong and a light wind.
Text on an object: a mug that reads MORNING
PromptClose-up of a white ceramic mug on a wooden table by a window, steam rising, the word "MORNING" printed in bold black letters on the mug, shallow depth of field, the camera slowly orbits the mug, quiet room tone.
Stylized 3D animation: a small robot waters a plant
PromptStylized 3D animation: a small round orange robot waters a potted plant on a windowsill while rain streaks down the glass, soft warm indoor light, the plant grows a new leaf, gentle piano music and the patter of rain.
| Test | Model | Settings | Credits | Run time | Result |
|---|---|---|---|---|---|
| Fast human motion: a breakdance spin on a wet street | Kling 4.0 | 5 s · 720p · 16:9 | 600 credits | 3 min 42 s | Done |
| Fast human motion: a breakdance spin on a wet street | Kling 3.0 | 5 s · 720p · 16:9 · with sound | 83 credits | 3 min 6 s | Done |
| Speech and lip sync: a barista talks to the camera | Kling 4.0 | 5 s · 720p · 16:9 | 600 credits | 3 min 19 s | Done |
| Speech and lip sync: a barista talks to the camera | Kling 3.0 | 5 s · 720p · 16:9 · with sound | 83 credits | 3 min 1 s | Done |
| Landscape and camera move: drone over a misty forest | Kling 4.0 | 5 s · 720p · 16:9 | 600 credits | 4 min 31 s | Done |
| Landscape and camera move: drone over a misty forest | Kling 3.0 | 5 s · 720p · 16:9 · with sound | 83 credits | 2 min 58 s | Done |
| Text on an object: a mug that reads MORNING | Kling 4.0 | 5 s · 720p · 16:9 | 600 credits | 5 min 10 s | Done |
| Stylized 3D animation: a small robot waters a plant | Kling 4.0 | 5 s · 720p · 16:9 | 600 credits | 4 min 9 s | Done |
- Fast motion (the breakdance spin). The dancer stayed in one piece through fast footwork around a hand planted in the puddles, while the camera circled and the neon reflections held steady. Water splashed where the hands hit the street.
- Speech and lip sync (the barista). The barista slid a paper cup across the counter in a wide shot, then the clip cut to a close-up for the line, spoken word for word: “Your oat latte is ready. Careful, it is hot.” The mouth moves in time with the words.
- Landscape and camera move (the drone shot). A steady forward glide over the pines toward a river winding through a misty valley, with sun rays in the fog. The hills and the river kept their shape for the whole move.
- Text on an object (the MORNING mug). MORNING came out spelled right in bold black letters, as asked, and stayed right while the camera moved around the mug and the handle came into view.
- Stylized animation (the robot). The robot watered the plant from a small can, then the clip cut to a close-up of the water and to a wide shot where a bright new leaf has appeared. In the first shot, a large blurred copy of the robot shows in the window behind it.
Photo to video
For the image-to-video test we made a still of a red fox in snow with Seedream 5.0 Pro on DER, then asked Kling 4.0 to bring it to life.
Photo to video
The input is a still we generated with Seedream 5.0 Pro. 5 seconds, 720p, with sound.
Photo to video: the fox from the still image
PromptThe fox turns its head toward the camera, blinks, then stands up and trots away into the birch forest as snowflakes keep falling. Soft wind and the crunch of snow.
Input

| Test | Model | Settings | Credits | Run time | Result |
|---|---|---|---|---|---|
| Photo to video: the fox from the still image | Kling 4.0 | 5 s · 720p | 600 credits | 5 min 17 s | Done |
The fox kept the look of the still, from its coat and face to the birch trees behind it. It looked at the camera, got up and trotted off to the left until it left the frame, so the clip ends on empty snow rather than with the fox heading into the trees.
Kling 4.0 or Kling 3.0?
The first three prompts also ran on Kling 3.0, at the same length and resolution and with sound turned on, so you can see what changed between versions.
Kling 4.0 followed the prompts more closely. On the dance, it kept the dancer up on hands and feet through fast footwork, while Kling 3.0 played it as a spin on the ground. Both baristas said the line word for word, but Kling 4.0 delivered all of it in a close-up, while Kling 3.0's barista spoke most of it with the face still out of frame. On the drone shot, Kling 3.0 framed the sunrise well, but its clip came back almost silent even with sound turned on.
On price, Kling 3.0 starts from 32 credits for a 3-second clip (≈ $0.32) and Kling 4.0 starts from 360 credits for a 3-second clip (≈ $3.60). In our runs a 5-second 720p clip with sound cost 83 credits on Kling 3.0 and 600 on Kling 4.0.
Choose Kling 4.0 when a character has to speak on camera or the clip has to match a detailed prompt. Kling 3.0 cost 83 credits for the same 5-second 720p clip with sound, about a seventh of Kling 4.0's 600, which makes it a good way to try out an idea first.
What a Kling 4.0 clip costs
These are the credits DER charged for each clip in this test:
| Clip | Length | Resolution | Credits |
|---|---|---|---|
| Text to video | 5 s | 720p | 600 |
| Image to video | 5 s | 720p | 600 |
Kling 4.0 is priced per second of video, so a longer clip costs more. For 1080p or any other length, the Studio shows the exact credits before you start. To compare Kling 4.0 with every other video model on DER, see pricing.
How to get better results with Kling 4.0
- Ask for one continuous shot if you want no cuts. Kling 4.0 cut to a close-up on its own in the barista and robot clips.
- Put the spoken line in quotes. Our barista said the quoted line word for word.
- Put on-screen words in quotes and describe the lettering. “MORNING” in bold black letters came out exactly as written.
- Name the camera move in plain words. “Aerial shot gliding forward” and “the camera slowly orbits” both came out as described.
- In image to video, give the direction from the camera's point of view, such as “away from the camera, into the trees”. Our fox went off to the side.
Try Kling 4.0
Open the Kling 4.0 model page for its settings and starting price, or go straight to text to video or image to video. The Kling family page lists every Kling version on DER.
How we tested
Six prompts: five text-to-video prompts and one image-to-video test from a still we made with Seedream 5.0 Pro on DER. Every Kling 4.0 clip is 5 seconds, 720p and 16:9; Kling 4.0 always adds sound. The first three prompts also ran on Kling 3.0 at 5 seconds and 720p, with sound turned on. Times run from starting the job to the finished clip. We checked the spoken line with speech-to-text.
Credits are what each run actually cost on November 4, 2026. Prices change: each model's page has the current one.