Kling O1 API

Kling O1 video generation model with image-to-video, video editing, and fast video editing variants. Supports 3-20 second videos with reference images for style-guided generation.

Prompt

Transform the video with cinematic color grading and smooth transitions.

Input Video

Upload a video for editing (6-20 seconds, max 100MB). Supports MP4, WebM, MOV, AVI formats.

Reference Images (Optional)

Upload up to 4 reference images for style guidance.

Parameters

Aspect Ratio: 16:9Keep Original Sound: Yes

Sample Result

72 (suggested: 2,000)

Upload a video for editing (max 100MB)

Click to upload or drag and drop

Supported formats: MP4, MOV
Maximum file size: 100MB; Maximum files: 1

Upload reference images

Click to upload or drag and drop

Supported formats: JPG, JPEG, PNG, WEBP
Maximum file size: 10MB; Maximum files: 4

0:00 / 0:00
Audio

History

Max 20 items

0 running · 0 completed

Your generation history will appear here

Kling O1 API for unified video generation and editing

Build modern video workflows with Kling O1. Use one API to create new clips from prompts, refine existing footage, and keep characters and scenes consistent across outputs for marketing, social content, and commerce.

Hero showcase of Kling O1 video feature 1

What can you build with the Kling O1 API?

Prompt-to-video storytelling

Turn short creative briefs into videos with Kling O1 and keep the same look across multiple outputs. This is useful for social campaigns, brand series, or episodic content where consistency matters more than one-off experimentation.

Storytelling showcase of Kling O1 video feature 2

Reference-driven edits

Use Kling O1 to refine or rework existing footage with instruction-based edits. Keep the core subject intact while adjusting style, lighting, or scene details so teams can iterate fast without a full re-shoot.

Editing showcase of Kling O1 video feature 3

Commercial content at scale

Kling O1 is positioned for production teams in advertising, e-commerce, and social media. Use it to generate variations, keep brand tone consistent, and deliver content at the pace required by modern channels.

Commerce showcase of Kling O1 video feature 4

Why teams choose Kling O1

Kling O1 emphasizes unified creation and editing with consistency across characters and scenes, which reduces rework and keeps creative pipelines predictable.

Unified multimodal workflow

Text, image, video, and subject inputs live in one model.

Consistency-first outputs

Maintain recognizable characters and scenes across clips.

Production-friendly focus

Built for film, social, ads, and commerce workflows.

How to integrate the Kling O1 API

A simple flow from input to production-ready video.

1

Choose inputs and mode

Select text, image, video, or subject inputs based on your workflow and desired output type.

2

Submit a generation task

Send your request with instructions and any references, then track the task until results are ready.

3

Review and iterate

Download results, compare variations, and reuse the same structure for fast iteration.

Core capabilities of the Kling O1 API

Unified video creation and editing in one model

Core

Unified multimodal engine

Kling O1 is introduced as a unified multimodal model that combines generation and editing in a single system. This allows teams to keep one integration while handling both new clip creation and edits across the same workflow.

Inputs

Text, image, video, and subject inputs

Public descriptions highlight that Kling O1 supports text, image, video, and subject inputs. This gives creators more ways to control outputs and reduces guesswork when consistent results are required.

Consistency

Consistency for characters and scenes

Kling O1 is positioned to address the consistency challenge in AI video generation. This helps teams keep character identity, props, and scene details aligned across multiple clips.

Workflow

Generation plus editing workflows

Instead of switching tools, Kling O1 brings generation and editing tasks into one engine. This is useful for marketing teams that need to create, then refine, without breaking continuity.

Use Cases

Commercial content use cases

The model is described as suitable for film, television, social media, advertising, and e-commerce workflows. That makes it a practical choice for teams building content at scale.

MVL

Multimodal visual language

Kling O1 is built on a multimodal visual language framework. This helps it interpret intent across text and visual references so outputs align more closely with creative direction.

Frequently Asked Questions

Everything you need to know about the product and billing.

The Kling O1 API provides access to Kling O1, which is described as a unified multimodal video model. It brings text, image, video, and subject inputs into a single engine for generation and editing. For teams, this means one integration can support both new clip creation and instruction-based edits, while aiming for consistent characters and scenes across outputs. Availability, limits, and pricing depend on your provider and region, so always confirm the current details in your EvoLink dashboard and API responses before building production workflows.
Public announcements state that Kling O1 integrates text, image, video, and subject inputs in one model. In practice, this gives you multiple control paths: you can guide video creation with a prompt, anchor results with references, or refine existing footage. Which inputs are available to your account can vary by endpoint and plan, so it is best to follow the input requirements in your API documentation and use your dashboard as the source of truth when you standardize pipelines.
Kling O1 is positioned as a unified system that covers both generation and editing tasks in one model. That means you can create new videos from prompts and also refine or transform existing footage without switching tools. This unified workflow is helpful for production teams because it reduces rework and keeps creative direction consistent. Exact editing tools and parameters depend on the provider, so check the current API docs for the supported controls in your region.
The launch announcement describes Kling O1 as addressing the consistency challenge in AI video generation, especially for characters and scenes. For teams, that means it can be easier to keep subject identity and visual style stable across multiple clips. Consistency still depends on the quality of your inputs and the clarity of your instructions, so adopt prompt templates and reference guides to keep output stable across campaigns and channels.
Kling O1 is presented as a unified creation tool for film, television, social media, advertising, and e-commerce. These are all environments where teams need many outputs that look and feel aligned. If you are building a marketing or commerce workflow, Kling O1 can help you generate variations, test creative options, and keep branding consistent across short-form video content.
Kling O1 is described as consolidating generation and editing tasks into a single engine, which reduces the need to stitch together multiple APIs for different stages of a video workflow. For developers, this simplifies integration and makes it easier to scale. Still, you should confirm which endpoints are exposed by your provider and how they map to specific tasks, since coverage can vary by region and plan.
Start with a clear subject and describe the action, mood, and setting in simple terms. If you are using reference inputs, keep them focused on the elements you want to preserve, such as a character or a product. Consistency improves when your prompt structure stays stable across runs, so create a shared prompt template that your team can reuse for campaigns and different markets.
Limits, pricing, and available input types are determined by your provider and region. Use your EvoLink dashboard and the API responses as the source of truth, and check the linked API documentation for the most current constraints. If you are planning a large campaign, validate throughput and concurrency with your provider before launch so you can scale without workflow delays.
POST
/v1/videos/generations

Edit Video (Fast)

Kling O1 Video Edit Fast (kling-o1-video-edit-fast) model transforms and edits videos with AI-powered style transfer.

Asynchronous processing mode, use the returned task ID to query status.

Generated video links are valid for 24 hours, please save them promptly.

Important Notes

  • Supports 6-20 second videos for editing.
  • Videos under 6 seconds are billed as 6 seconds.
  • Videos over 20 seconds will be truncated to 20 seconds.
  • Maximum file size: 100MB per video.
  • Supported formats: MP4, WebM, MOV, AVI.
  • Optional: Up to 4 reference images for style guidance.

Request Parameters

modelstringRequiredDefault: kling-o1-video-edit-fast

Video editing model name.

Examplekling-o1-video-edit-fast
promptstringRequired

Prompt describing how to edit and transform the video.

Notes
  • Limited to 2000 tokens
ExampleTransform the video with cinematic color grading and smooth transitions.
video_urlsarrayRequired

Input video URL list for video editing.

Notes
  • Required: exactly 1 video
  • Maximum: 1 video per request
  • Max size: 100MB per video
  • Formats: .mp4, .mov
  • Duration: 6-20 seconds (longer videos will be truncated)
  • URLs must be directly accessible by the server
Example["https://example.com/input-video.mp4"]
image_urlsarrayOptional

Reference image URL list for style guidance.

Notes
  • Optional
  • Maximum: 4 images per request
  • Max size: 10MB per image
  • Formats: .jpg, .jpeg, .png, .webp
  • Use reference images to guide the AI in understanding your desired style
Example["https://example.com/reference1.jpg", "https://example.com/reference2.jpg"]
keep_original_soundbooleanOptionalDefault: true

Whether to keep the original audio from the input video.

ValueDescription
truePreserve original audio track
falseGenerate new audio or mute
Exampletrue
aspect_ratiostringOptionalDefault: 16:9

Output video aspect ratio.

ValueDescription
16:9Landscape video
9:16Portrait video
1:1Square video
Example'16:9'
callback_urlstringOptional

HTTPS callback address after task completion.

Notes
  • Triggered on completion, failure, or cancellation
  • Sent after billing confirmation
  • HTTPS only, no internal IPs
  • Max length: 2048 chars
  • Timeout: 10s, Max 3 retries
Examplehttps://your-domain.com/webhooks/video-task-completed

Request Example

{
  "model": "kling-o1-video-edit-fast",
  "prompt": "Transform the video with cinematic color grading and smooth transitions.",
  "video_urls": [
    "https://example.com/input-video.mp4"
  ],
  "image_urls": [
    "https://example.com/reference-style.jpg"
  ],
  "keep_original_sound": true,
  "aspect_ratio": "16:9",
  "callback_url": "https://your-domain.com/webhooks/callback"
}

Response Example

{
  "created": 1757169743,
  "id": "task-unified-1757169743-abc123",
  "model": "kling-o1-video-edit-fast",
  "object": "video.generation.task",
  "progress": 0,
  "status": "pending",
  "task_info": {
    "can_cancel": true,
    "estimated_time": 120,
    "video_duration": 10
  },
  "type": "video",
  "usage": {
    "billing_rule": "per_second",
    "credits_reserved": 68,
    "user_group": "default"
  }
}
Kling O1 API: Unified Video Creation & Editing | EvoLink