Kling 4.0. One Kling take, up to 30 seconds.
Kuaishou's Kling 4.0 renders a silent clip of 3 to 30 seconds from a prompt, an opening frame, or a first and last frame, in 21:9, 16:9, 1:1 or 9:16. On Treza it runs through fal, in the Video Generation node.
- 3 to 30 second clips
- 4 aspect ratios
- First and last frame control
- 720p or 1080p
- No audio
- Runs on fal
Kling 4.0 is the version of Kuaishou's Kling video model that follows Kling 3.0. Treza runs it through fal, because OpenRouter carries Kling 3.0 and Kling O1 but not 4.0. It takes a prompt, a first frame, or a first and a last frame, and renders any whole length from 3 to 30 seconds at 720p or 1080p, in 21:9, 16:9, 1:1 or 9:16. A clip that opens on a frame takes that image's shape. It makes no sound.
fal listed it on 2026-11-12 at 15.7 cents a second at 720p and 23.5 cents at 1080p. Treza renders 720p unless you pick 1080p on the node and bills exactly the length you set, so a 10 second 720p clip is $1.57 of provider cost. Treza quotes the list price plus its credit markup before a run, and a failed generation costs nothing. It runs on the canvas, in the Video Generation node, and through any pipeline you publish as an API. Chat does not run it; there, ask for Kling 3.0 Pro.
Up to 30 seconds in one take
Any whole second from 3 to 30, twice the 15 second ceiling of Kling 3.0 Pro, so a full scene or a whole social beat comes out of one generation instead of a stitch.
First and last frame
Open on your own image, or pin both ends and the clip travels between them. With a frame wired, the clip follows the image's shape rather than the aspect ratio setting.
Four shapes, two tiers
21:9 for a cinematic wide, 16:9, 1:1, and 9:16 for Shorts and Reels. 720p by default, 1080p when the clip will play large.
Picture only
Kling 4.0 has no audio setting and returns a silent clip. Add narration with a Text to Speech node or a bed from a music node, and mix them in the Sequence node.
What to make with Kling 4.0
Real prompts, and what comes back. Every one runs as a node on the Treza canvas.
“A 30 second single take: a cyclist rides a coastal road from dawn into full sun, tracking shot, 21:9”
One long cinematic take, with no stitching.
“Start on this product photo (attached): the camera circles it slowly as the light warms”
Image to video that opens on your frame and keeps its shape.
“First frame: an empty studio. Last frame: the same studio set for a shoot (both attached)”
A clip that travels between two frames you supply.
“A dancer turns through falling confetti, tracking shot, 9:16, 12 seconds”
A vertical beat for Shorts and Reels.
“Last week's Kling 3.0 Pro prompt, re-rendered on Kling 4.0 at 25 seconds”
A take longer than 3.0 Pro can render.
“Rain runs down a window at night, city lights blurred behind, 1:1, 1080p”
A square clip at the higher tier, ready for a music bed.
Kling 4.0 is one node on a canvas
Chain it with language, image, and video models, then publish the whole pipeline as an API. See how a full AI video pipeline fits together.
- Step 01
Draft the brief
A language model node expands a one-line idea into a detailed, on-brand prompt before it ever reaches the model.
- Step 02
Generate with Kling 4.0
Run variants in parallel, compare against other models on the same prompt, and keep every result in run history.
- Step 03
Publish it as an API
The pipeline becomes a versioned endpoint. Your product calls it with the typed /invoke API or any OpenAI SDK.
Kling 4.0 as an API, priced per clip
The Kling 4.0 API
Build a pipeline with Kling 4.0 on a node and publish it. It becomes a versioned HTTP endpoint: the typed /invoke route enqueues a run and returns a run id to poll, and an OpenAI-compatible route works with SDKs you already use. Agents connect over MCP the same way. Your prompt engineering ships inside the pipeline, so every caller gets your quality, and swapping the model later does not change the API your product calls.
Kling 4.0 pricing
Prepaid credits, no subscription. Each clip is charged at Kling 4.0's own rate from your balance, shown before you run, and a failed generation costs nothing. Credit packs start at $5 and never expire, which suits spiky production schedules. Run history records the cost of every generation, per node, so spend is auditable rather than a surprise.
Kling 4.0, answered
What is Kling 4.0?
Kling 4.0 is the version of Kuaishou's Kling video model that follows 3.0. It turns a prompt, a first frame, or a first and last frame into a silent clip of 3 to 30 seconds at 720p or 1080p. Treza runs it through fal, in the Video Generation node, because OpenRouter does not carry it.
What does Kling 4.0 cost?
fal listed it on 2026-11-12 at 15.7 cents a second at 720p and 23.5 cents at 1080p, so a 10 second clip is $1.57 or $2.35 of provider cost. Treza bills the length you set at that rate, quotes it plus its credit markup before a run, and a failed generation costs nothing.
What is the difference between Kling 4.0 and Kling 3.0 Pro?
Length, shape, sound and price. Kling 4.0 renders 3 to 30 seconds in 21:9, 16:9, 1:1 or 9:16, at 720p or 1080p, with no sound. Kling 3.0 Pro renders 3 to 15 seconds in 16:9, 9:16 or 1:1, with sound on or off. fal lists Kling 4.0 at 15.7 cents a second at 720p; OpenRouter lists 3.0 Pro at 11.2 cents silent and 16.8 with sound. Chat runs 3.0 Pro, and Kling 4.0 runs on the canvas.
How is Kling 4.0 Flash different?
Flash is fal's lower-cost Kling 4.0 endpoint: 720p only, 3 to 20 seconds, and a first frame but no last frame, at 9.4 cents a second against 15.7 for Kling 4.0 at 720p. Draft on Flash, then re-render the keeper on Kling 4.0 when it needs 1080p, a longer take, or an end frame.
Can I use Kling 4.0 in chat?
It runs on the canvas, in the Video Generation node. In chat, ask for Kling 3.0 Pro.
How much does Kling 4.0 cost on Treza?
Treza uses prepaid credits with no subscription. Each clip is charged at Kling 4.0's rate from your balance, and only successful generations are charged. Credit packs start at $5 and never expire.
Can I call Kling 4.0 through an API?
Yes. Build a pipeline that uses Kling 4.0 and publish it as a versioned HTTP endpoint. Call the typed /invoke API from your product, or point any OpenAI SDK at the OpenAI-compatible endpoint. If a better model ships later, swap it on the node without changing the API your product calls.
Your first video is one prompt away.
Generate your first video, image, or draft today. Prepaid credits, no subscription.