Kling O1 API

Text to VideoImage to Video

Kling's reasoning model for complex prompts. Superior understanding and creative interpretation.

·Features·FAQ

Documentation

Prompt *

Text description for the video generation

Upload images/videos above, then type @ in your prompt to reference them

Reference images

Up to 7; reference as @image_1 … @image_7

0/7

Click or drag & dropPNG, JPG, WEBP, GIF · Max 100MB

Elements JSON

Optional IMAGE-only elements (see docs). Shared pool of 7 with image_urls.

Start Image

Upload a starting image (optional)

0/1

Click or drag & dropPNG, JPG, WEBP, GIF · Max 100MB

End frame

Optional ending frame

0/1

Click or drag & dropPNG, JPG, WEBP, GIF · Max 100MB

Reference video

Optional reference clip; use @video_1 in prompt

0/1

Click or drag & dropMP4, WEBM, MOV · Max 100MB

Video mode

Applies when a reference video is used

Keep reference video audio

Keep audio from the reference video

Duration

Video duration in seconds

Aspect Ratio

Output video aspect ratio

Mode

Output quality mode

Count

Number of outputs (1-4)

Output

Your generated video will appear here

Features

What Kling O1 API offers

Text-to-video driven by a required prompt with duration from 3 to 10 seconds

Up to 7 reference images plus optional IMAGE-only elements JSON (shared pool of 7)

Optional start and end frames, optional reference video with Reference or Transform behavior

Optional keep audio from the reference clip

Aspect ratio 16:9, 9:16, or 1:1

Standard (720p) or Pro (1080p) output

Optional batch of 1 to 4 videos per request

REST API with JSON request and response bodies

Use cases

Built for

Primary

Grounded scenes - Reference images and elements so props and people stay recognizable

Continuity edits - Start and end frames to bookend a motion beat

Shot guidance - Reference clip in Reference or Transform mode to mirror timing or style

Multi-take review - Count greater than 1 when you want several options per brief

Dialogue-led spots - Keep audio from stock reference when useful

FAQ

About Kling O1 API

Kling O1 is a prompt-led video model with rich conditioning: reference images, IMAGE-only elements, optional frames, optional reference video, configurable duration from 3 to 10 seconds, and no multi-shot mode in this schema.

Follow the docs for @image_1 tags, named elements, @video_1 when a reference clip is present, and frame fields when you supply bookend images.

O1 caps single-clip duration at 10 seconds and omits Auto aspect or the 3.0 Omni omni bundle. Kling 3.0 adds multi-shot timelines, longer 15 second singles, audio generation defaults, and Omni adds Auto aspect and different reference limits.

You can ask for 1 to 4 outputs in one job to compare takes without re-uploading assets.

O1 targets structured prompting with references. Master is still the dedicated Pro-only path inside the older 2.1 family. Pick the tool that matches whether you need reference bundles or a simpler 2.1 Pro pipeline.