What is multi-shot in Kling 3.0?

Multi-shot lets one request return a connected sequence of up to five shots, each with its own prompt and duration (1–12s). Set multi_shots to true and provide multi_prompt. Kling 3.0 keeps characters and scenes consistent across the cuts — ideal for ads, storyboards, and short narratives.

How do I do image-to-video with Kling 3.0?

Include image_urls: the first entry is the start frame, and an optional second entry is the end frame (Kling 3.0 interpolates between them). URLs must be public HTTPS (no base64). Aspect ratio auto-adapts to your images, so the aspect_ratio field is ignored when frames are supplied.

How does Kling 3.0 pricing work?

Credit-based: 1 credit equals $0.001 USD. Kling 3.0 bills per second, dimensioned by mode (std 720p / pro 1080p / 4K) and whether audio is on. Total cost is the per-second rate times the video length; multi-shot bills the sum of shot durations. Failed and rejected requests are refunded. See the pricing card above for current rates.

What durations and aspect ratios does Kling 3.0 support?

Single-shot runs 3–15 seconds; multi-shot supports up to five shots of 1–12 seconds each. Aspect ratios are 16:9, 9:16, and 1:1 (default 16:9), auto-adapted when you supply frame images. Resolution is std 720p, pro 1080p, or 4K via the mode field.

Yes. You submit a Kling 3.0 task and receive a task_id, then poll /api/v1/tasks/:id until it is completed. This is the same task flow as every other reApi model, and the API returns a video URL you should mirror to your own storage if you need it long term.

Where is the full API reference for Kling 3.0?

The complete parameter list, modes, validation rules, error codes, and runnable cURL / Python / Node / Go examples for Kling 3.0 live at /docs/kling-3-0.

Is Kling 3.0 free or paid?

Paid. Completed generations are charged at the rates in this page's pricing table. Eligible new accounts may receive promotional test credits; availability and amount are shown at signup or in the dashboard. Technical or provider failures are refunded automatically, except where this page documents a charged output-moderation block.

Does Kling 3.0 require a monthly subscription?

No. Pricing is pay-as-you-go: $1 buys 1,000 credits and credits never expire. Cost scales with what you generate rather than with a recurring fee.

How do I get a Kling 3.0 API key?

Sign up and create a key in the dashboard, no card required. Pass it as a Bearer token in the Authorization header.

Kling 3.0 — Multi-Shot AI Video with Native Audio

Q: Does Kling 3.0 generate audio?

Yes. Turn sound on and Kling 3.0 generates synchronized audio with the video — multilingual dialogue with lip sync, dialects and accents, plus ambient sound. Audio tiers are priced above silent ones; you pay for sound only when you enable it.

Kling 3.0 is Kuaishou's flagship video model: write a prompt for text-to-video, animate a still with image-to-video, or direct a multi-shot sequence — each shot with its own prompt, duration, and camera work. It generates native multilingual audio with lip sync, up to 4K, and you call Kling 3.0 through reApi's unified API.

Highlights: 3–15 s · std / pro / 4K · multi-shot prompts
Best for: Cinematic shots, multi-scene stories
Input: Text, plus up to 2 images
Output: Video, optional sound

Input

Prompt

Up to 2500 characters · required for single-shot

Frame images

Drag images here or click to upload

0/2 files

Public HTTPS URLs only (no base64) · up to 2 (first / last frame)

Duration6

3–15 seconds · billed per second

Aspect ratio

16:9 / 9:16 / 1:1 · default 16:9

Mode

std 720p · pro 1080p · 4K · price scales with the tier

Native audio

Off by default · audio tiers cost more per second

Multi-shot

Pair with Multi-shot prompts below

Multi-shot prompts

JSON: up to 5 shots of prompt + duration · 1–12s each

Elements

JSON: up to 3 elements, each with 2–4 image URLs · reference with @name in the prompt

Estimated cost~$0.594594 credits

Result

Try one of these prompts

What is the Kling 3.0 API?

Kling 3.0 is Kuaishou's flagship video model, exposed on reAPI as one async endpoint covering text-to-video, image-to-video from a first frame with an optional last frame, and multi-shot sequences of up to 5 shots that each carry their own prompt and duration, plus up to 3 reusable elements built from 2–4 reference images. Every request renders 3–15 seconds in 16:9, 9:16, or 1:1 at std, pro, or 4K quality with optional native audio, billed per second of output at the rate for the tier you pick.

What you can build with this model

Real-world workflows and production use cases you can build and ship with this model.

Direct a multi-shot sequence

The headline upgrade in Kling 3.0 is multi-shot. Hand it up to five shots — each with its own prompt, duration, and camera direction — and Kling 3.0 returns one connected sequence with consistent characters and scenes across cuts. It is built for storyboards, ads, and short narratives where a single static clip is not enough, all from one request and one task ID.

Open the playground

Animate a still with image-to-video

Pass a start frame — and optionally an end frame — and Kling 3.0 animates the still into smooth motion, interpolating between the two when both are supplied. Aspect ratio auto-adapts to your images. It is the fastest way to turn a product shot, a character render, or a key frame into a moving clip without re-describing the whole scene in text.

Video with native audio

Turn on sound and Kling 3.0 generates synchronized audio with the video — multilingual dialogue with lip sync, dialects and accents, plus ambient sound — instead of a separate audio pass. Pair it with multi-shot prompts and you get talking characters across cuts, which makes Kling 3.0 a one-call tool for narrated shorts and dialogue scenes.

Pricing

Credit-based — 1 credit = $0.001 USD. Pay only for completed generations.

First test cost

$0.385385 credits

One 5-second sample at the cheapest tier.

Testing budget guide

Add credits

$10

≈ 25 tests

$50

≈ 129 tests

$100

≈ 259 tests

Model	Category	Price
std 720p · no audio	Per second	$0.077 77 credits · 1 second
std 720p · audio	Per second	$0.11 110 credits · 1 second
pro 1080p · no audio	Per second	$0.099 99 credits · 1 second
pro 1080p · audio	Per second	$0.149 149 credits · 1 second
4K	Per second	$0.369 369 credits · 1 second

Why reAPI

Multi-shot storytelling in one call

Most video models give you a single clip. Kling 3.0 directs up to five shots — distinct prompts, durations, and camera moves — into one coherent sequence. You skip the stitch-and-sync busywork, and characters stay consistent across cuts through the same request to /api/v1/videos/generations.

Native audio, not a separate pass

Kling 3.0 generates synchronized audio with the picture — multilingual dialogue, lip sync, dialects, and ambient sound. Turn `sound` on and the clip comes back voiced. It is billed per second, with audio tiers priced above silent ones, so you pay for sound only when you ask for it.

Async-first, unified

Submit a Kling 3.0 task, get a task_id back, poll until it completes. 1 credit equals $0.001 USD and you pay per second × tier, so cost is predictable. The JSON matches the OpenAI generations contract, so adding Kling 3.0 is a model-id change, not a new integration.

Kling 3.0 vs Veo 3

Both generate premium video with audio from one API. Kling 3.0 leans into multi-shot direction, multilingual audio, and flexible duration with per-second pricing. Here is how the two compare on publicly documented behavior.

Capability

Kling 3.0 on reAPI

Veo 3

Multi-shot sequences

Up to five shots in one request — each with its own prompt, duration, and camera direction.

Single-clip generation; sequences are stitched from multiple requests.

Native audio

Multilingual dialogue with lip sync, dialects, and ambient sound, generated with the video.

Native audio including dialogue and sound effects.

Image-to-video

Start frame plus optional end frame, with aspect ratio auto-adapted.

Image-to-video from a single reference image.

Resolution

std 720p, pro 1080p, or 4K via mode.

Up to 1080p / 4K depending on tier.

Duration

3–15s single-shot; multi-shot up to five 1–12s shots.

Fixed short clip lengths.

Pricing model

Per second × tier × audio; pay-as-you-go credits, no subscription.

Per-second pricing that scales with resolution and audio.

Comparison reflects publicly documented behavior at the time of writing. Model behavior and pricing can change; check the pricing card above and the API docs for current values.

Integrate Kling 3.0 in three steps

step 01
Create an API key
Sign up at reApi and grab an API key. Free signup credits cover your first Kling 3.0 clips — enough to test text-to-video and image-to-video before you top up.
Open
step 02
Submit a generation
POST to /api/v1/videos/generations with model: kling-3-0, a prompt, and optional duration / aspect_ratio / mode / sound / image_urls. Kling 3.0 returns a task_id immediately.
Open
step 03
Poll the result
GET /api/v1/tasks/:id until status is completed. The Kling 3.0 response carries the video URL; mirror it to your own storage if you need it long term, since generated links expire.
Open

docs/api/kling-3-0

API reference

Drop-in code and the full parameter table.

View full reference

curl -X POST https://reapi.ai/api/v1/videos/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-3-0",
    "prompt": "A soft anime-style girl in a field, warm sunset glow, smooth loop",
    "duration": 6,
    "aspect_ratio": "16:9",
    "mode": "pro"
  }'

Parameter validation failed

Rejected synchronously with the failing field named. Check enum values, ranges, and URL fields — nothing is charged.

Authentication or balance issue

401 means a missing or invalid Bearer key; 402 means the reserve exceeds your balance. Manage keys and credits in the console.

Content or material rejected

Reference material and outputs pass automated moderation. Rejected tasks fail with a clear error and a full refund.

Task failed or timed out

A task that reaches a failed state is never charged — the reserve refunds automatically. Keep the task id and retry when ready.

Frequently asked questions

Common questions about this model.

Kling 3.0 is Kuaishou's flagship video model. It does text-to-video and image-to-video, generates native multilingual audio with lip sync, and — new in 3.0 — directs multi-shot cinematic sequences. On reApi, Kling 3.0 ships on the unified /api/v1/videos/generations endpoint under the model id kling-3-0.