Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Kuaishou (Kling) / Kling Video

Kling 3.0 Turbo

Text and image video generation with model-specific duration, resolution and audio controls.

Video generation
availableClosed weightsVideo generation
Official documentation (opens in a new tab)Reviewed 13 Oct 2026
7 of 21 specification fields collected
Output resolution720p, 1080pSee specification notes below
Clip duration3–15 secondsSee specification notes below
Input billingSee ratesKling AI API
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected
Maximum output
Not collected
Input modalities
Text, ImageSource for Input modalities (opens in a new tab)
Output modalities
Video, AudioSource for Output modalities (opens in a new tab)
Reasoning
Not collected
Tool calling
Not collected
Structured output
Not collected
Streaming
Not collected
Batch processing
Not collected
Fine-tuning
Not collected
Knowledge cutoff
Not collected
Release date
Not collected
Parameters
Not collected
Architecture
Not collected
Languages
Not collected
Context caching
Not collected
Video duration
3–15 secondsSource for Video duration (opens in a new tab)
Output resolutions
720p, 1080pSource for Output resolutions (opens in a new tab)
Maximum prompt characters
2,500 charactersSource for Maximum prompt characters (opens in a new tab)
Aspect ratios
16:9, 9:16, 1:1Source for Aspect ratios (opens in a new tab)
Generation modes
1–6 shots; each at least one second; shot durations sum to the requested duration; up to 512 characters per shot.Source for Generation modes (opens in a new tab)

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Kling AI API

Provider source (opens in a new tab)
API model ID
kling-3.0-turbo

Published capability-map identifier. The current route selects this model; the new request body does not take model_name.

Access
Authenticated Kling API account and eligible balance/package. New and legacy authentication requirements vary by route.
Regions
Public examples use api-singapore.klingai.com; this hostname does not establish contractual data residency.
API endpoint URL
https://api-singapore.klingai.com/image-to-video/kling-3.0-turboSource for API endpoint URL (opens in a new tab)
Request limits
Concurrency depends on the account’s active package and is shared across API keys. Image tasks consume slots by image count; querying tasks does not consume concurrency.Source for Request limits (opens in a new tab)
Output link expiry
Generated assets are cleared after 30 days; save output URLs promptly.Source for Output link expiry (opens in a new tab)

Artifact availability only, not a training-use policy or a contractual zero-retention guarantee.

Request mode
Asynchronous task submission with polling or callbackSource for Request mode (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 13 Oct 2026
CategoryUsageRateUnit
VideoNative audio · 720p1$0.112second
VideoNative audio · 1080p2$0.14second

Rate conditions

  1. Official public API USD list price. Native audio; 720p output. Package discounts and account terms may differ; not web-app credits.
  2. Official public API USD list price. Native audio; 1080p output. Package discounts and account terms may differ; not web-app credits.
Pricing source (opens in a new tab)

Data & enterprise terms

Retention, training use, regional processing and enterprise terms have not been summarized for this endpoint.

Kling AI API

Provider source (opens in a new tab)
API model ID
kling-3.0-turbo

Published capability-map identifier. The current route selects this model; the new request body does not take model_name.

Access
Authenticated Kling API account and eligible balance/package. New and legacy authentication requirements vary by route.
Regions
Public examples use api-singapore.klingai.com; this hostname does not establish contractual data residency.
API endpoint URL
https://api-singapore.klingai.com/text-to-video/kling-3.0-turboSource for API endpoint URL (opens in a new tab)
Request limits
Concurrency depends on the account’s active package and is shared across API keys. Image tasks consume slots by image count; querying tasks does not consume concurrency.Source for Request limits (opens in a new tab)
Output link expiry
Generated assets are cleared after 30 days; save output URLs promptly.Source for Output link expiry (opens in a new tab)

Artifact availability only, not a training-use policy or a contractual zero-retention guarantee.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 13 Oct 2026
CategoryUsageRateUnit
VideoNative audio · 720p1$0.112second
VideoNative audio · 1080p2$0.14second

Rate conditions

  1. Official public API USD list price. Native audio; 720p output. Package discounts and account terms may differ; not web-app credits.
  2. Official public API USD list price. Native audio; 1080p output. Package discounts and account terms may differ; not web-app credits.
Pricing source (opens in a new tab)

Data & enterprise terms

Retention, training use, regional processing and enterprise terms have not been summarized for this endpoint.

Alibaba Model Studio · Beijing · Kling AI

Provider source (opens in a new tab)
API model ID
kling/kling-v3-turbo-video-generation
Access
Activate the Kling provider in Model Studio, then use a Beijing key. POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; Authorization: Bearer <key>, X-DashScope-Async: enable, Content-Type: application/json. Poll GET https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/tasks/{task_id}.
Regions
China (Beijing)
Input modalities
Text, ImageSource for Input modalities (opens in a new tab)
Output modalities
Video, AudioSource for Output modalities (opens in a new tab)

Audio behavior is route/model dependent. This is not a standalone TTS model.

Maximum prompt characters
2,500Source for Maximum prompt characters (opens in a new tab)

Top-level prompt; custom shots have a separate 512-character prompt limit.

Languages
Chinese, EnglishSource for Languages (opens in a new tab)

Documented prompt languages, not an exhaustive text-rendering or speech-language guarantee.

Input image formats
JPEG, JPG, PNGSource for Input image formats (opens in a new tab)

Alpha channel is unsupported; HTTP(S) reference URLs are documented, not Base64.

Image input limits
300–8,000 pixels per side; aspect ratio 1:2.5 through 2.5:1; at most 10 MB per image. No alpha channel.Source for Image input limits (opens in a new tab)
Input media transport
Public HTTP(S) media URLs. No Base64 contract established for this hosted route.Source for Input media transport (opens in a new tab)
Asynchronous generation
SupportedSource for Asynchronous generation (opens in a new tab)
Synchronous generation
Not supportedSource for Synchronous generation (opens in a new tab)
Streaming
Not collected

An asynchronous task API does not establish streaming support or non-support.

Request mode
Async HTTP; X-DashScope-Async: enable is required. Use a Beijing key and workspace endpoint, then poll the task resource.Source for Request mode (opens in a new tab)
Task ID lifetime
Task IDs can be queried for 24 hours; expired/unknown tasks return UNKNOWN.Source for Task ID lifetime (opens in a new tab)
Output link expiry
Generated download URLs are valid for 30 days. This is not a guarantee about deletion of provider copies or input retention.Source for Output link expiry (opens in a new tab)
Task-status query rate
Task query default: 20 requests per second; a fifteen-second polling interval is suggested.Source for Task-status query rate (opens in a new tab)
Published request quota
300 requests per minuteSource for Published request quota (opens in a new tab)
Task submission rate
5 requests per secondSource for Task submission rate (opens in a new tab)

RPS and RPM are separately published constraints; their burst behavior is not treated as equivalent.

Concurrent requests
Ten running tasks shared across the five Kling image/video models per Alibaba Model Studio API keySource for Concurrent requests (opens in a new tab)
Free usage allowance
No free quotaSource for Free usage allowance (opens in a new tab)
Published availability regions
China (Beijing)Source for Published availability regions (opens in a new tab)
Hosted service availability
2026-08-27Source for Hosted service availability (opens in a new tab)

Alibaba-hosted availability date, not the original native model release date.

Visible watermark
false by default. true additionally returns a version with a fixed bottom-right Kling AI watermark.Source for Visible watermark (opens in a new tab)
Tool calling
Not supportedSource for Tool calling (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Structured output
Not supportedSource for Structured output (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Context caching
Not supportedSource for Context caching (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Batch processing
Not supportedSource for Batch processing (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Fine-tuning
Not supportedSource for Fine-tuning (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Search grounding
Not supportedSource for Search grounding (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Prefix completion
Not supportedSource for Prefix completion (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Console model playground
Not supportedSource for Console model playground (opens in a new tab)

Hosted model-card declaration only; does not overwrite native-provider capability facts.

Negative prompt limit
Not collected

negative_prompt is ignored by this hosted model; put negative descriptions in the main prompt.

Negative prompt support
Not supportedSource for Negative prompt support (opens in a new tab)

The field is ignored, not treated as an independent prompt.

Output formats
MP4, H.264Source for Output formats (opens in a new tab)
Videos per request
1Source for Videos per request (opens in a new tab)

Returned video_count is fixed at one; multi-shot generation is still one output video.

Output resolutions
720p, 1080pSource for Output resolutions (opens in a new tab)

mode=pro (1080p) is default; std=720p; 4k is unavailable for Turbo.

Video duration
Integer 3–15 seconds; default 5.Source for Video duration (opens in a new tab)
Aspect ratios
16:9, 9:16, 1:1Source for Aspect ratios (opens in a new tab)

Explicit aspect_ratio required for text/reference generation. First-frame tasks inherit that image ratio; editing inherits input video ratio.

Reference images
One first frame only. Last frame and subject entities are unsupported.Source for Reference images (opens in a new tab)
Multi-shot generation
multi_shot=false by default. When true, shot_type is required: intelligence or customize. Custom mode uses 1–6 indexed shots, each prompt ≤512 characters (excess is truncated), duration integer 1–parameters.duration.Source for Multi-shot generation (opens in a new tab)

Do not infer an undocumented total-shot-duration validation rule.

Audio generation
Always generates synchronized audio; audio=false is ignored.Source for Audio generation (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation720p · synchronized audio1CN¥0.80second of output video
Video generation1080p · synchronized audio2CN¥1.00second of output video

Rate conditions

  1. Original Beijing CNY price; promotions/taxes excluded. No free quota. Turbo always generates audio; the audio flag is ignored. Applies only to kling/kling-v3-turbo-video-generation through Alibaba, not the native Kling API.
  2. Original Beijing CNY price; promotions/taxes excluded. No free quota. Turbo always generates audio; no 4K or separate silent price is published. Applies only to kling/kling-v3-turbo-video-generation through Alibaba, not the native Kling API.
Pricing source (opens in a new tab)

Data & enterprise terms

Kling AI is the inference provider; Alibaba supplies this regional access contract. Hosted input retention, training use and inference data residency are not independently established. Generated URLs expire after 30 days.

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

Kling AI API

kling-3.0-turbo
Not verified
How it works
Not collected
Minimum reusable input
Not collected
Lifetime & refresh
Not collected
How to enable it
Not collected
Usage fields
Not collected
Billing rules
Not collected
Limits & data boundaries
Not collected

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Kling AI API

kling-3.0-turbo
Not verified
How it works
Not collected
Minimum reusable input
Not collected
Lifetime & refresh
Not collected
How to enable it
Not collected
Usage fields
Not collected
Billing rules
Not collected
Limits & data boundaries
Not collected

No verified cache rate for this endpoint. An empty rate is not a zero charge.

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Inference from the exact named production model or managed service in Kuaishou’s official API catalog: no downloadable production weights or open-weight license were established for this identity. Original public-weight Kolors and research derivatives are excluded; public demo code is not evidence of production weights.

Classification evidence (opens in a new tab)

Lifecycle dates

API availability announced17 Jun 2026Source (opens in a new tab)
Alibaba Model Studio Beijing availability27 Aug 2026Source (opens in a new tab)

Sources reviewed