Kling 2.6 Pro: Realistic Image-to-Video with Sound Sync on playground and API

kling/kling-2-6/pro/image-to-video

Transform images into cinematic videos with synchronized sound, lifelike motion, and expressive voices, delivering fast, high-fidelity storytelling for creators, marketers, and developers.

Prompt *

Image *

URL of the image to be used for the video. Accepted file types: jpg, jpeg, png, webp, gif, avif

Duration

The duration of the generated video in seconds.

Negative Prompt

A negative prompt to exclude undesired qualities (e.g., blur, distortion, low quality).

Generate Audio

Whether to generate native audio for the video. Supports Chinese and English voice output. Other languages are automatically translated to English. For English speech, use lowercase letters; for acronyms or proper nouns, use uppercase.

Idle

The rate is $0.07 per second without audio, and $0.14 per second with audio.

Introduction to Kling 2.6 Pro Features

Kuaishou Technology's Kling 2.6 Pro turns text or a single reference image into 5 to 10 second 1080p cinematic video, priced at $0.07 per second without audio or $0.14 per second with native audio, with simultaneous audio-visual generation. Trading silent renders and manual sound design for one-pass, lip-synced dialogue, SFX, ambience, and cinematic motion coherence, Kling 2.6 Pro streamlines short-form production, preserves character identity from a reference image, and eliminates separate audio post for agencies, social marketers, and VFX previsualization teams on RunComfy. For developers, Kling 2.6 Pro on RunComfy can be used both in the browser and via an HTTP API, so you don’t need to host or scale the model yourself.
Ideal for: High-Conversion Video Ads | Cinematic Previsualization | Character-Consistent Image-to-Video Social Clips

Examples Created Using Kling 2.6 Pro

Kling 2.6 Pro on X: Latest Discussions

Model Overview

Provider: Kuaishou Technology (Kling AI)
Task: image-to-video
Max Resolution/Duration: 1080p / up to 10s
Summary: Kling 2.6 Pro generates short cinematic videos from a single image with synchronized native audio, delivering coherent motion and consistent character identity. It supports English and Chinese speech, respects concise scene/action/camera cues, and outputs 5–10 second clips in multiple aspect ratios. For technical artists, Kling 2.6 Pro emphasizes high-fidelity lighting, temporal stability, and reliable prompt adherence.

Key Capabilities

Native audio-visual co-generation with lip-sync

Kling 2.6 Pro produces visuals and audio (dialogue, SFX, ambience, music) in one pass, delivering frame-accurate synchronization and natural lip movement for English and Chinese.
This reduces post-production steps and ensures dialogue timing and sound staging match camera motion and on-screen action.

Reference image consistency for cinematic motion

In image-to-video mode, Kling 2.6 Pro preserves identity, style, and facial fidelity from a single still image while adding natural motion.
The model maintains subject consistency across frames, with improved lighting realism, textures, and facial-motion coherence.

Predictable results from concise, structured prompts

Kling 2.6 Pro responds strongly to focused prompts describing subject, action, camera, lighting, and optional dialogue/ambience.
Technical users get reproducible 1080p results for 5–10 second sequences, with selectable aspect ratios (16:9, 9:16, 1:1) and optional audio control.

Input Parameters

Core Inputs

Parameter	Type	Default/Range	Description
prompt	string	""	Required. Describe subject, action, camera, lighting, and optional dialogue/ambience for audio.
image_url	string (image URI)	""	Required. Publicly accessible URL to the reference image used for image-to-video generation.
negative_prompt	string	"blur, distort, and low quality"	Optional. List attributes to avoid (e.g., artifacts, unwanted styles, lighting).

Dimensions & Audio Settings

Parameter	Type	Default/Range	Description
duration	integer	5 or 10	Clip length in seconds. Choose 5s or 10s.
generate_audio	boolean	true	Toggle native audio generation. Supports English/Chinese speech; other languages auto-translate to English. For English, use lowercase; use uppercase for acronyms/proper nouns.

How Kling 2.6 Pro compares to other models

Vs earlier Kling versions: Compared to prior releases (e.g., 2.5 or 1.x), Kling 2.6 Pro delivers one-pass audio-visual generation with lip-sync, better motion coherence, improved facial fidelity, and stronger prompt adherence. Ideal when synchronized sound and cinematic polish are critical.
Vs Seedance 1.0 Pro: Compared to Seedance 1.0 Pro, Kling 2.6 Pro emphasizes integrated native audio with dialogue/SFX in the same render and robust prompt-to-motion execution. Ideal for scenes where dialogue timing and emotional delivery must align tightly with visuals.
Vs Wan 2.5: Compared to Wan 2.5, Kling 2.6 Pro focuses on 1080p output and consistent subject identity from a single reference image, with reliable audio toggle behavior and strong adherence to concise prompts. Ideal when cinematic lighting and tight character consistency are top priorities within a 5–10 second, full-HD workflow.
Ideal Use Case: Choose Kling 2.6 Pro for short-form cinematic beats, social ads, and previsualization that demand character consistency and synchronized native audio without separate sound design passes.

API Integration

Developers can integrate Kling 2.6 Pro via the RunComfy API using standard HTTP requests with simple JSON payloads. Authentication, job submission, and result polling follow familiar REST patterns, enabling quick pipeline adoption in production or toolchains.

Note: API Endpoint for Kling 2.6 Pro

Official resources and licensing

Official Website/Paper: https://app.klingai.com/global/quickstart/klingai-video-26-audio-user-guide
License: Proprietary model accessible via supported platforms. Commercial use is governed by platform/provider terms and may require a separate agreement.

Explore Related Capabilities

If you do not have a reference image and want to generate directly from text, use Kling 2.6 Pro — Text-to-Video, which is optimized for prompt-driven scene creation and native audio.

Related Playgrounds

hailuo-2-3/pro/text-to-video

AI-powered video creation tool offering 1080p motion and natural expression for precise, artistic storytelling.

seedance-1-0/lite/image-to-video

Make fast, realistic videos from text or images at a low cost.

one-to-all-animation/14b

Transforms static characters into smooth motion clips for flexible creative workflows

sync/lipsync/v2

Create lifelike synced videos from voices or images with precise motion and creative control.

hailuo-02/pro/image-to-video

Animate an image into a smooth 6s video with Hailuo 02 Pro.

infinite-talk/fast/video-to-video

AI model for dynamic dubbing and expressive video creation from voice or footage.

Frequently Asked Questions

What type of license does Kling 2.6 Pro use, and can I use its image-to-video outputs commercially?

Kling 2.6 Pro follows a Non-Commercial or restricted Open RAIL-style license depending on your access channel. Using Kling 2.6 Pro image-to-video outputs through RunComfy does not change the original licensing terms — you must comply with Kuaishou Technology’s official policies when using generated content for commercial distribution.

What are the technical limitations of Kling 2.6 Pro when generating image-to-video clips?

Kling 2.6 Pro currently supports up to 1080p resolution across common aspect ratios (16:9, 9:16, 1:1). Prompt inputs are limited to around 1,000 tokens, and image-to-video sessions allow 1–2 reference images per render. Exceeding these constraints can cause warnings or degraded fidelity.

How can I transition from testing Kling 2.6 Pro in the Playground to production use via RunComfy’s API?

You can start with the Kling 2.6 Pro Web Playground to test your image-to-video prompts, then move to RunComfy’s API using your API key. The API mirrors Playground behavior but supports automated scaling, enabling you to integrate Kling 2.6 Pro directly into commercial or enterprise workflows.

What performance improvements does Kling 2.6 Pro offer over earlier Kling versions for image-to-video applications?

Kling 2.6 Pro introduces better facial motion, smoother transitions, built-in audio generation, and more accurate prompt interpretation than Kling 2.5. Its image-to-video results show stronger character consistency and lighting realism, bringing it closer to cinematic-grade output quality.

Can I disable audio output when using Kling 2.6 Pro for image-to-video generation?

Yes, Kling 2.6 Pro provides a toggle to disable audio, allowing silent image-to-video clips when native sound is unnecessary. This feature is useful for projects where you plan to add voiceover or sound design later in post-production.

What should developers know about latency and throughput when calling Kling 2.6 Pro via RunComfy API?

The average latency for Kling 2.6 Pro image-to-video generation is approximately 10–20 seconds per 5-second clip, depending on scene complexity and system load. RunComfy API requests queue intelligently to maintain stable concurrency across high-traffic periods.

Does RunComfy’s use of Kling 2.6 Pro grant me unrestricted rights to distribute generated image-to-video content?

No. RunComfy provides access to Kling 2.6 Pro under Kuaishou’s defined license. Even when generating image-to-video content through RunComfy, users must comply with the original model’s licensing terms, including any limitations around redistribution or commercial monetization.

How does Kling 2.6 Pro handle aspect ratio selection for image-to-video outputs?

Kling 2.6 Pro supports 16:9, 9:16, and 1:1 aspect ratios during image-to-video generation. Selecting the ratio before rendering ensures optimal composition and framing, particularly for platforms like YouTube (16:9) or TikTok (9:16).

Is Kling 2.6 Pro suitable for enterprise-scale image-to-video production?

Yes. Kling 2.6 Pro is optimized for scalability via RunComfy’s cloud API, with GPU resource pooling for large teams. Its image-to-video capabilities enable automated marketing content or storytelling applications, but commercial use still requires adherence to the model’s licensing conditions.

Support

Video Models/Tools

Image Models

Legal

RunComfy

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

Kling 2.6 Pro: Realistic Image-to-Video with Sound Sync on playground and API | RunComfy