ComfyUI Node: Image2Image Sampler v1 (Searge)

Image2Image Sampler v1 (Searge):

The SeargeSDXLImage2ImageSampler is a powerful node designed to transform an existing image into a new one by leveraging advanced sampling techniques. This node is particularly useful for AI artists who want to apply high-resolution fixes and refine their images with precision. It utilizes a combination of base and refiner models to enhance the image quality, ensuring that the final output is both detailed and aesthetically pleasing. The node's primary goal is to provide a seamless and efficient way to perform image-to-image transformations, making it an essential tool for creative professionals looking to elevate their artwork.

Image2Image Sampler v1 (Searge) Input Parameters:

data

This parameter holds the initial data required for the sampling process. If not provided, the node will initialize it as an empty dictionary. It serves as the foundation for the subsequent parameters and ensures that the necessary information is available for the sampling process.

sampler_input

This parameter contains the specific inputs needed for the sampler. If not provided, it will be retrieved from the data parameter. It includes various sub-parameters such as base and refiner models, positive and negative prompts, latent images, and more. These inputs are crucial for guiding the sampling process and determining the final output.

base_model

This parameter specifies the base model to be used for the initial image transformation. It is a critical component that influences the overall style and quality of the generated image. The base model serves as the starting point for the image refinement process.

base_positive

This parameter contains the positive prompts for the base model. Positive prompts guide the model towards desired features and characteristics in the generated image, ensuring that the output aligns with the artist's vision.

base_negative

This parameter contains the negative prompts for the base model. Negative prompts help the model avoid unwanted features and characteristics, refining the image to better match the desired outcome.

refiner_model

This parameter specifies the refiner model to be used for enhancing the image quality. The refiner model works in conjunction with the base model to add finer details and improve the overall resolution of the image.

refiner_positive

This parameter contains the positive prompts for the refiner model. Similar to the base positive prompts, these guide the refiner model towards desired features, ensuring that the final image is detailed and high-quality.

refiner_negative

This parameter contains the negative prompts for the refiner model. These prompts help the refiner model avoid unwanted features, further refining the image to meet the artist's expectations.

latent_image

This parameter holds the latent image data, which serves as the input for the sampling process. The latent image is a representation of the initial image in a compressed form, allowing the models to process and transform it efficiently.

noise_seed

This parameter specifies the seed for generating noise during the sampling process. The noise seed ensures that the transformations are reproducible and consistent. The default value is 4815162342.

steps

This parameter determines the number of steps to be taken during the sampling process. More steps generally result in higher quality images but may increase the processing time. The default value is 25.

cfg

This parameter stands for Classifier-Free Guidance and controls the strength of the guidance during the sampling process. A higher value results in stronger guidance, leading to more pronounced features. The default value is 7.0.

sampler_name

This parameter specifies the name of the sampler to be used. Different samplers may produce varying results, and the choice of sampler can influence the style and quality of the final image. The default value is "dpmpp_2m".

scheduler

This parameter determines the scheduling method for the sampling process. The scheduler controls the order and timing of the steps, affecting the overall efficiency and quality of the transformation. The default value is "karras".

base_ratio

This parameter specifies the ratio of the base model's influence in the final image. A higher base ratio means that the base model has a stronger impact on the output. The default value is 0.8.

denoise

This parameter controls the level of denoising applied during the sampling process. Higher denoise values result in smoother images with fewer artifacts. The default value is 1.0.

cfg_method

This parameter specifies the method for dynamic Classifier-Free Guidance. It allows for more flexible and adaptive guidance during the sampling process, enhancing the overall quality of the image.

dynamic_base_cfg

This parameter sets the dynamic base Classifier-Free Guidance value. It provides additional control over the guidance strength, allowing for more precise adjustments. The default value is 0.0.

Image2Image Sampler v1 (Searge) Output Parameters:

data

This output parameter contains the final data after the sampling process. It includes the transformed image and any additional information generated during the process. The data parameter is essential for accessing the final output and further processing or saving the image.

out_denoised

This output parameter holds the denoised version of the final image. It provides a cleaner and smoother version of the output, free from noise and artifacts. The denoised image is ideal for final presentation and use in various creative projects.

Image2Image Sampler v1 (Searge) Usage Tips:

• Experiment with different base and refiner models to achieve various artistic styles and effects.
• Adjust the number of steps to balance between image quality and processing time.
• Use positive and negative prompts effectively to guide the models towards desired features and avoid unwanted characteristics.
• Fine-tune the Classifier-Free Guidance (CFG) value to control the strength of the guidance and achieve the desired level of detail.

Image2Image Sampler v1 (Searge) Common Errors and Solutions:

"Sampler input not found"

• Explanation: This error occurs when the sampler_input parameter is not provided and cannot be retrieved from the data parameter.
• Solution: Ensure that the sampler_input parameter is correctly provided or included in the data parameter.

"Base model not specified"

• Explanation: This error occurs when the base_model parameter is missing or not specified.
• Solution: Provide a valid base model in the sampler_input parameter to proceed with the sampling process.

"Refiner model not specified"

• Explanation: This error occurs when the refiner_model parameter is missing or not specified.
• Solution: Provide a valid refiner model in the sampler_input parameter to enhance the image quality.

"Invalid noise seed"

• Explanation: This error occurs when the noise_seed parameter is not a valid integer.
• Solution: Ensure that the noise_seed parameter is a valid integer value.

"Steps parameter out of range"

• Explanation: This error occurs when the steps parameter is set to a value outside the acceptable range.
• Solution: Adjust the steps parameter to a value within the acceptable range, typically between 1 and 100.