Style Model Apply Advanced(StyleModelApplyAdvanced)
This node receives a base conditioning input, a loaded style model, and CLIP visual features extracted from a reference image.
Style Model Apply Advanced
The Style Model Apply Advanced node receives a base conditioning, a loaded style model, and CLIP visual features extracted from a reference image.
Style Model Apply Advanced is a node designed for advanced style control, capable of injecting a specific visual style into the generation process with adjustable intensity. Its core function is to modify and enhance a given text conditioning using a precomputed style model and CLIP visual features.
Node Function
This node receives a base conditioning, a loaded style model, and CLIP visual features extracted from a reference image. Its internal mechanism involves calling the style model to generate a style conditioning tensor based on the CLIP visual features, then scaling this tensor according to the user-defined intensity value. Finally, it concatenates the scaled style conditioning with the input base conditioning along a specific dimension, thereby producing a new conditioning that incorporates the specified style intensity.
Node Parameter Description - Style Model Apply Advanced
Connection Inputs (Inputs)
| Parameter Name | Data Type | Required | Default Value | Range/Options | Description |
|---|---|---|---|---|---|
conditioning | CONDITIONING | Yes | - | - | The input base text conditioning, which will be fused with the style conditioning. Type is CONDITIONING. |
style_model | STYLE_MODEL | Yes | - | - | The loaded pre-trained style model, used to generate the style conditioning. Type is STYLE_MODEL. |
clip_vision_output | CLIP_VISION_OUTPUT | Yes | - | - | The CLIP visual features extracted from the reference image, serving as the basis for style generation. Type is CLIP_VISION_OUTPUT. |
Control Parameters (Parameters)
| Parameter Name | Data Type | Required | Default Value | Range/Options | Description |
|---|---|---|---|---|---|
strength | FLOAT | Yes | 1.0 | -10.0 - 10.0 (Step: 0.001) | The coefficient controlling the intensity of style application, ranging from -10.0 to 10.0 with a step size of 0.001. |
Output
| Parameter Name | Data Type | Description |
|---|---|---|
| CONDITIONING | CONDITIONING | The new conditioning information fused with the specified style intensity. |
Usage Scenario
When building a workflow to generate images in a specific artistic style (e.g., Van Gogh's style), you can first process the reference image using the CLIP Vision Encode node, then load the corresponding style model using the Load Style Model node. Next, convert the text prompt into base conditioning via the CLIP Text Encode node, and finally connect it to this node. You can precisely control the prominence of the style in the final image by adjusting the strength parameter.
Notes
The strength parameter has a wide range (-10.0 to 10.0). Setting it to a negative value may produce unpredictable or opposite style effects, so caution is advised. Additionally, the quality of the style model and the accuracy of the CLIP visual features directly impact the final result.
Style Model Apply Advanced Node Source Code Link
The Style Model Apply Advanced node is from the ComfyUI-KJNodes node package.
Comments
Sign in with GitHub to join the discussion.