Visit ComfyUI Online for ready-to-use ComfyUI environment
Specialized node for sampling in AI art, leveraging multiple models simultaneously for diverse outputs.
KSamplerXY| KSampler XY 🍌 is a specialized node designed to facilitate the sampling process in AI art generation by leveraging multiple models simultaneously. This node extends the capabilities of the standard KSampler by allowing you to input an array of models (model_xy
) and generate samples from each model in a coordinated manner. The primary benefit of using KSamplerXY| KSampler XY 🍌 is its ability to combine the strengths of different models, potentially leading to more diverse and high-quality outputs. The node is particularly useful for artists looking to experiment with various model configurations and achieve unique artistic effects by blending the outputs of multiple models.
This parameter accepts an array of models (XY_MODEL
) that will be used for sampling. Each model in the array contributes to the final output, allowing for a combination of different model characteristics.
The seed
parameter is an integer that initializes the random number generator, ensuring reproducibility of the results. The default value is 0, with a minimum of 0 and a maximum of 0xffffffffffffffff.
The steps
parameter defines the number of sampling steps to be performed. More steps generally lead to higher quality outputs but require more computational resources. The default value is 20, with a minimum of 1 and a maximum of 10000.
The cfg
(Classifier-Free Guidance) parameter is a float that controls the strength of the guidance. Higher values result in outputs that more closely follow the conditioning inputs. The default value is 8.0, with a range from 0.0 to 100.0, adjustable in steps of 0.1 and rounded to 0.01.
This parameter specifies the name of the sampler to be used. It accepts values from the predefined list of samplers available in comfy.samplers.KSampler.SAMPLERS
.
The scheduler
parameter determines the scheduling strategy for the sampling process. It accepts values from the predefined list of schedulers available in comfy.samplers.KSampler.SCHEDULERS
.
The positive
parameter is used for positive conditioning, guiding the model towards desired features in the output. It accepts a CONDITIONING
type input.
The negative
parameter is used for negative conditioning, guiding the model away from undesired features in the output. It accepts a CONDITIONING
type input.
The latent_image
parameter provides the initial latent image to be used as the starting point for the sampling process. It accepts a LATENT
type input.
The denoise
parameter is a float that controls the amount of denoising applied during the sampling process. The default value is 1.0, with a range from 0.0 to 1.0, adjustable in steps of 0.01.
The samples
output parameter contains the final generated samples after combining the outputs from all the models in the model_xy
array. The samples are concatenated into a single tensor, providing a unified output that reflects the combined characteristics of the input models.
model_xy
array to achieve unique artistic effects.cfg
parameter to fine-tune the balance between following the conditioning inputs and allowing for creative variations.seed
parameter to ensure reproducibility of your results, especially when you find a configuration that produces desirable outputs.steps
parameter for higher quality outputs, but be mindful of the increased computational requirements.model_xy
array are not of the expected XY_MODEL
type.model_xy
array are correctly specified and compatible with the XY_MODEL
type.seed
parameter is set to a value outside the acceptable range.seed
value is within the range of 0 to 0xffffffffffffffff.steps
parameter is set to a value outside the acceptable range.steps
value is within the range of 1 to 10000.cfg
parameter is set to a value outside the acceptable range.cfg
value is within the range of 0.0 to 100.0.denoise
parameter is set to a value outside the acceptable range.denoise
value is within the range of 0.0 to 1.0.© Copyright 2024 RunComfy. All Rights Reserved.