Unifically LogoUnificAlly
Model logo

Kling 2.6 API

Text to VideoImage to Video

Latest Kling model with enhanced dynamics and aesthetics. Supports text-to-video and image-to-video.

Click or drag & dropPNG, JPG, WEBP, GIF · Max 100MB
Click or drag & dropPNG, JPG, WEBP, GIF · Max 100MB
Generate Audio
Generate audio for the output video
Output

Your generated video will appear here

Features

What Kling 2.6 API offers

Text-to-video with a required prompt, or image-to-video with a start image and optional prompt text
Optional end frame image when you want to suggest how the clip resolves
Duration 5 or 10 seconds
Aspect ratio 16:9, 9:16, or 1:1 when generating from text (aspect applies to text-to-video in the API)
Standard (720p) or Pro (1080p) output
Optional generated audio for the output clip
REST API with JSON request and response bodies

Use cases

Built for

Primary

Social clips - Text-to-video for 5 or 10 second hooks in standard widescreen or vertical formats

#2

Product loops - Start from a pack shot image and prompt the motion you want

#3

Storyboards - Rough motion between a supplied start and optional end frame

#4

Ads and explainers - Pro mode when you need sharper 1080p for paid placements

#5

Internal reviews - Shareable MP4 previews before committing to a heavier model

FAQ

About Kling 2.6 API

Kling 2.6 generates MP4 video from either a text prompt or a starting image. You can set duration to 5 or 10 seconds, choose Standard or Pro quality, optionally add audio generation, and attach an optional end image.

For text-to-video the prompt is required. For image-to-video the start image enables the run and the prompt is optional text guidance.

Aspect ratio selection applies to text-to-video generation in this integration. Image-to-video centers on the supplied frames plus duration and quality.

Kling 3.0 adds multi-shot timelines, longer 3 to 15 second single clips, and different defaults for audio. Kling 2.6 stays at 5 or 10 seconds with a simpler parameter surface.

Use the Unifically REST endpoints for generate and status with a JSON body built from your prompt, frames, duration, mode, and options. The model docs describe fields and examples.