ComfyUI Node: Pre CFG automatic scale

Pre CFG automatic scale:

The Pre CFG automatic scale node is designed to dynamically adjust the Classifier-Free Guidance (CFG) scale during the denoising process in AI art generation. This node helps in fine-tuning the balance between the conditional and unconditional outputs, ensuring that the generated images adhere closely to the desired artistic style while maintaining a high level of detail and coherence. By automatically scaling the CFG, this node enhances the overall quality of the generated images, making it a valuable tool for AI artists looking to achieve precise control over their creative outputs.

Pre CFG automatic scale Input Parameters:

model

This parameter specifies the model to be used for the automatic scaling process. It is essential as it defines the framework within which the scaling adjustments will be applied.

maximum_scale

This parameter sets the upper limit for the CFG scale. It determines the maximum intensity of guidance that can be applied during the denoising process. The default value is 80, with a minimum of 0.0 and a maximum of 1000.0. Adjusting this value can significantly impact the adherence to the conditional input.

minimum_scale

This parameter sets the lower limit for the CFG scale. It defines the minimum level of guidance to be applied. The default value is 4.5, with a minimum of 0.0 and a maximum of 10.0. Lower values can result in more creative freedom but may reduce adherence to the conditional input.

strength

This parameter controls the intensity of the scaling effect. The default value is 0.5, with a range from 0.0 to 10.0. Higher values increase the impact of the automatic scaling adjustments, while lower values reduce it.

end_at_sigma

This parameter specifies the sigma value at which the scaling adjustments should stop. The default value is 0.28, with a range from 0.0 to 1000.0. This helps in controlling the point at which the guidance scaling ceases, ensuring a smooth transition in the denoising process.

converging_scales

This boolean parameter determines whether the scales should converge during the process. The default value is True. When enabled, it ensures that the conditional and unconditional outputs gradually align, enhancing the coherence of the final image.

invert_mask

This boolean parameter allows for the inversion of the input mask. The default value is False. When enabled, it inverts the areas affected by the scaling, providing an alternative approach to modifying the image.

input_mask (optional)

This parameter accepts a mask input that defines the areas of the image to be modified by the node. The mask helps in selectively applying the CFG scale adjustments, allowing for targeted enhancements.

input_latent (optional)

This parameter accepts a latent input that sets the maximum scale allowed for seeking similarity. It helps in defining the scope of the scaling adjustments, ensuring that the modifications stay within the desired range.

Pre CFG automatic scale Output Parameters:

model

The output model is the modified version of the input model with the applied automatic CFG scaling adjustments. This model incorporates the dynamic scaling changes, resulting in improved image quality and adherence to the desired artistic style.

Pre CFG automatic scale Usage Tips:

• Experiment with different maximum_scale and minimum_scale values to find the optimal balance between adherence to the conditional input and creative freedom.
• Use the strength parameter to fine-tune the intensity of the scaling effect, adjusting it based on the complexity of the desired output.
• Enable the converging_scales parameter to ensure a smooth and coherent transition between the conditional and unconditional outputs, especially for highly detailed images.

Pre CFG automatic scale Common Errors and Solutions:

"Invalid model input"

• Explanation: The model input provided is not compatible with the node.
• Solution: Ensure that the model input is correctly specified and compatible with the Pre CFG automatic scale node.

"Scale value out of range"

• Explanation: The specified maximum_scale or minimum_scale values are outside the allowed range.
• Solution: Adjust the scale values to be within the specified range (0.0 to 1000.0 for maximum_scale and 0.0 to 10.0 for minimum_scale).

"Invalid sigma value"

• Explanation: The end_at_sigma value is not within the acceptable range.
• Solution: Ensure that the end_at_sigma value is within the range of 0.0 to 1000.0.

"Mask input not provided"

• Explanation: The input_mask parameter is required but not provided.
• Solution: Provide a valid mask input to the input_mask parameter to define the areas to be modified by the node.