Kling 3.0 Video and Image Generation in ComfyUI

Learn how to use Kling 3.0 models in ComfyUI for multi-shot video generation, subject consistency, multilingual audio, and native text rendering

Kling 3.0 is one of the most advanced multi-modal generation systems, now available in ComfyUI via Partner Nodes. This release includes Kling Video 3.0, Kling Video 3.0 Omni, Kling Image 3.0, and Kling Image 3.0 Omni, bringing video, image, audio-visual, and narrative generation capabilities directly into your node-based workflows.

What Kling 3.0 is good at

• Multi-shot generation: Generate multiple shots with duration control in a single generation
• Locked subject consistency: Maintain character and object identity across camera motion and scene evolution
• Native audio with character awareness: Multilingual dialogue with precise lip sync and facial expression alignment
• Native text rendering: Clear, structured on-screen text for ads, UI scenes, and branded content

Example outputs

Text-to-image generation with the Kling 3.0 model:

Available models

Model	Type	Description
Kling Video 3.0	Video	Standard video generation with multi-shot support
Kling Video 3.0 Omni	Video	Advanced video generation with audio-visual capabilities
Kling Image 3.0	Image	High-quality image generation
Kling Image 3.0 Omni	Image	Multi-modal image generation with enhanced features

Learn more

For detailed examples, prompts, and video demonstrations of Kling 3.0’s capabilities, check out our blog post:

Kling 3.0 Models Are Now Available

Read the full announcement with video examples, prompt tips, and feature deep-dives.

To use the Partner Nodes, you need to ensure that you are logged in properly and using a permitted network environment. Please refer to the Partner Nodes Overview section of the documentation to understand the specific requirements for using the Partner Nodes.

Local users

Make sure your ComfyUI is updated.

• Download ComfyUI
• Update Guide

Workflows in this guide can be found in the Workflow Templates. If you can’t find them in the template, your ComfyUI may be outdated.

If nodes are missing when loading a workflow, possible reasons:

1. You are not using the latest ComfyUI version (Nightly version)
2. Some nodes failed to import at startup

Cloud users

• Cloud will update after ComfyUI stable release.

So, if you find any core node missing in this document, it might be because the new core nodes have not yet been released in the latest stable version. Please wait for the next stable release.

Kling 3.0: Text to Image

Generate an image from a text prompt using the Kling 3.0 model.

Run on Comfy Cloud

Open in Comfy Cloud

Download Workflow

Download JSON or search “Kling 3.0: Text to Image” in Template Library

Example output

Kling 3.0: Video Generation

Input text or image prompts to generate 15-second video sequences with multi-shot compositions, complex camera movements, and consistent subjects.

Run on Comfy Cloud

Open in Comfy Cloud

Download Workflow

Download JSON or search “Kling 3.0: Video Generation” in Template Library

Input materials

Upload this file to the matching LoadImage node:

boxing.png

LoadImage node 1 · boxing.png

Kling 3.0: First Last Frame to Video

Input a first and last frame image to generate a continuous video sequence with multi-shot generation and precise element control.

Run on Comfy Cloud

Open in Comfy Cloud

Download Workflow

Download JSON or search “Kling 3.0: First Last Frame to Video” in Template Library

Input materials

Upload these files to the matching LoadImage nodes:

goldenfish_in_bag.png

LoadImage node 1 · goldenfish_in_bag.png

man_in_daydream.png

LoadImage node 13 · man_in_daydream.png

Kling 3.0: Motion Control

Apply the motion of a reference video to a character image. Upload a character image plus a motion reference video and the model drives your character through the action in the clip while keeping their identity. This workflow runs the Kling 3.0 model, which improves facial consistency over previous versions.

Run on Comfy Cloud

Open in Comfy Cloud

Download Workflow

Download JSON or search “Kling 3.0: Motion Control” in Template Library

Input materials

Upload these files to the matching LoadImage and LoadVideo nodes:

future_robot.png

LoadImage node 2 · future_robot.png

boxing_man.mp4

LoadVideo node 3 · boxing_man.mp4

Motion control settings

Set character_orientation to video when the camera and framing should follow the motion clip, which allows reference clips up to 30 seconds, or to image when the character keeps the framing of the still image, which limits the clip to 10 seconds. Keep mode on pro for final output and switch to std for cheaper tests. keep_original_sound carries the clip’s audio into the result, and model switches back to kling-v2-6 for the previous generation.

This workflow uses the same Kling Motion Control node as the Kling 2.6 Motion Control example, so the input tips on that page apply here too.

Legacy Kling models (EOL)

End of life (EOL): The following legacy Kling models and workflow templates are scheduled for retirement on September 15, 2026: Kling V1.5, V1.6, V2.1, V2.1 Master, and the Kolors Virtual Try-On API. Related Partner Nodes remain in ComfyUI but these model options will stop working after the retirement date.

Retired workflow templates and replacements

Removed template	Suggested replacement
api_kling_i2v (V2.1 Master)	api_kling_o3_i2v or api_kling2_6_i2v
api_kling_effects (V1.6)	Kling 3.0 effect workflows (api_kling_v3_*)
api_kling_flf (V1.6)	api_kling_v3_flf2v or api_kling_o3_flf2v
api_kling_dual_characters_effects (V1.6)	Kling 3.0 effect workflows (api_kling_v3_*)

For virtual try-on, use Flux Virtual Try-On (api_flux_vto) or other supported alternatives. FLUX VTO and Gemini-based fashion templates are not affected by this retirement.