> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
> Use this file to discover all available pages before exploring further.

# KlingImageGenerationNode - ComfyUI Built-in Node Documentation

> Complete documentation for the KlingImageGenerationNode node in ComfyUI. Learn its inputs, outputs, parameters and usage.

> This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! [Edit on GitHub](https://github.com/Comfy-Org/embedded-docs/blob/main/comfyui_embedded_docs/docs/KlingImageGenerationNode/en.md)

Kling Image Generation Node generates images from text prompts with the option to use a reference image for guidance. It creates one or more images based on your text description and reference settings, then returns the generated images as output.

## Inputs

| Parameter         | Data Type | Required | Range                                                                                     | Description                                                                                                 |
| ----------------- | --------- | -------- | ----------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
| `prompt`          | STRING    | Yes      | -                                                                                         | Positive text prompt                                                                                        |
| `negative_prompt` | STRING    | Yes      | -                                                                                         | Negative text prompt                                                                                        |
| `image_type`      | COMBO     | Yes      | `"subject_reference"`<br />`"style_reference"`                                            | Image reference type selection (advanced). Required when a reference image is provided.                     |
| `image_fidelity`  | FLOAT     | Yes      | 0.0 - 1.0                                                                                 | Reference intensity for user-uploaded images (default: 0.5, advanced)                                       |
| `human_fidelity`  | FLOAT     | Yes      | 0.0 - 1.0                                                                                 | Subject reference similarity (default: 0.45, advanced)                                                      |
| `model_name`      | COMBO     | Yes      | `"kling-v3"`<br />`"kling-v2"`<br />`"kling-v1-5"`                                        | Model selection for image generation (default: "kling-v3")                                                  |
| `aspect_ratio`    | COMBO     | Yes      | `"16:9"`<br />`"9:16"`<br />`"1:1"`<br />`"4:3"`<br />`"3:4"`<br />`"21:9"`<br />`"9:21"` | Aspect ratio for generated images (default: "16:9")                                                         |
| `n`               | INT       | Yes      | 1 - 9                                                                                     | Number of generated images (default: 1)                                                                     |
| `image`           | IMAGE     | No       | -                                                                                         | Optional reference image                                                                                    |
| `seed`            | INT       | No       | 0 - 2147483647                                                                            | Seed controls whether the node should re-run; results are non-deterministic regardless of seed (default: 0) |

**Parameter Constraints:**

* The `image` parameter is optional. When a reference image is provided, the `image_type` parameter must be set to either `"subject_reference"` or `"style_reference"`.
* When no reference image is provided, the `image_type`, `image_fidelity`, and `human_fidelity` parameters are not used.
* Prompt and negative prompt have a maximum length of 500 characters.
* The `seed` parameter is optional and does not guarantee deterministic results.

## Outputs

| Output Name | Data Type | Description                                      |
| ----------- | --------- | ------------------------------------------------ |
| `output`    | IMAGE     | Generated image(s) based on the input parameters |

***

**Source fingerprint (SHA-256):** `3c029920c966ac9a3588a790e83e1350021a0864a02677258a25af46a0770dce`