ComfyUI Node: KSampler Sequence (v2)

KSampler Sequence (v2):

KSamplerSeq2 is a sophisticated node designed to facilitate the generation of image sequences using AI models. It leverages advanced sampling techniques to produce high-quality latent images, which can be further refined through iterative processes. This node is particularly beneficial for AI artists looking to create smooth transitions and coherent sequences in their artwork. By utilizing various parameters, KSamplerSeq2 allows for fine-tuning of the sampling process, enabling the creation of unique and visually appealing sequences. The node supports features such as latent interpolation, conditioning slerp, and noise injection, providing a versatile toolset for creative experimentation.

KSampler Sequence (v2) Input Parameters:

model

This parameter specifies the AI model to be used for generating the image sequences. The model is the core component that interprets the input parameters and produces the latent images.

seed

The seed parameter is an integer value that initializes the random number generator used in the sampling process. It ensures reproducibility of the generated sequences. The default value is 0, with a minimum of 0 and a maximum of 0xffffffffffffffff.

seed_mode_seq

This parameter determines the mode of seed variation across the sequence. Options include "increment", "decrement", "random", and "fixed". Each mode affects the sequence's randomness and coherence differently.

alternate_values

A boolean parameter that, when enabled, alternates certain values during the sequence generation. This can introduce variability and enhance the visual diversity of the output. The default value is True.

steps

The number of steps to be used in the sampling process. More steps generally lead to higher quality images but increase computation time. The default value is 20, with a minimum of 1 and a maximum of 10000.

cfg

The classifier-free guidance scale, which influences the trade-off between image fidelity and diversity. Higher values lead to more detailed images. The default value is 8.0, with a range from 0.0 to 100.0.

sampler_name

Specifies the name of the sampler to be used. Different samplers can produce varying results, and this parameter allows you to choose the one that best fits your artistic needs.

scheduler

Defines the scheduling strategy for the sampling process. Different schedulers can affect the timing and progression of the sampling steps.

sequence_loop_count

The number of times the sequence generation loop is executed. Higher values can produce longer and more complex sequences. The default value is 20, with a range from 1 to 1024.

positive_seq

Conditioning sequences that positively influence the generated images. These sequences guide the model towards desired features and characteristics.

negative_seq

Conditioning sequences that negatively influence the generated images. These sequences help the model avoid unwanted features and characteristics.

use_conditioning_slerp

A boolean parameter that enables spherical linear interpolation (slerp) between conditioning sequences. This can create smoother transitions between different conditions. The default value is False.

cond_slerp_strength

The strength of the conditioning slerp, determining how much influence the interpolation has on the final output. The default value is 0.5, with a range from 0.0 to 1.0.

latent_image

The initial latent image to be used as a starting point for the sequence generation. This image serves as the base for further refinements.

use_latent_interpolation

A boolean parameter that enables interpolation between latent images. This can create smooth transitions and enhance the coherence of the sequence. The default value is False.

latent_interpolation_mode

Specifies the mode of latent interpolation. Options include "Blend", "Slerp", and "Cosine Interp". Each mode affects the transition style differently.

latent_interp_strength

The strength of the latent interpolation, determining how much influence the interpolation has on the final output. The default value is 0.5, with a range from 0.0 to 1.0.

denoise_start

The initial denoising strength, which affects the amount of noise reduction applied at the beginning of the sequence. The default value is 1.0, with a range from 0.0 to 1.0.

denoise_seq

The denoising strength applied during the sequence generation. This parameter helps control the noise level throughout the process. The default value is 0.5, with a range from 0.0 to 1.0.

unsample_latents

A boolean parameter that enables the unsampling of latent images. This can introduce additional variability and enhance the visual diversity of the output. The default value is False.

alternate_mode

A boolean parameter that, when enabled, alternates certain modes during the sequence generation. This can introduce variability and enhance the visual diversity of the output. The default value is False.

inject_noise

A boolean parameter that enables the injection of noise into the latent images. This can create interesting visual effects and enhance the artistic quality of the output. The default value is True.

noise_strength

The strength of the noise to be injected, determining how much influence the noise has on the final output. The default value is 0.1, with a range from 0.0 to 1.0.

denoise_sine

A boolean parameter that enables sine-based denoising. This can create smooth and periodic noise reduction effects. The default value is True.

denoise_max

The maximum denoising strength, which affects the amount of noise reduction applied at the peak of the sine wave. The default value is 0.9, with a range from 0.0 to 1.0.

seed_keying

A boolean parameter that enables seed keying, which can introduce periodic variations in the seed value. This can create interesting visual effects and enhance the artistic quality of the output. The default value is True.

seed_keying_mode

Specifies the mode of seed keying. Options include "sine" and "modulo". Each mode affects the periodic variations in the seed value differently. The default value is "sine".

seed_divisor

The divisor used in the "modulo" seed keying mode, determining the frequency of seed variations. The default value is 4.

KSampler Sequence (v2) Output Parameters:

samples

The primary output of the node, which consists of the generated latent images. These images are the result of the sampling process and can be further refined or used as final outputs.

KSampler Sequence (v2) Usage Tips:

• Experiment with different seed modes to achieve varying levels of randomness and coherence in your sequences.
• Utilize the conditioning slerp and latent interpolation features to create smooth transitions and enhance the visual quality of your sequences.
• Adjust the denoise parameters to control the noise level and achieve the desired artistic effect.

KSampler Sequence (v2) Common Errors and Solutions:

"Invalid seed value"

• Explanation: The seed value provided is outside the acceptable range.
• Solution: Ensure the seed value is within the range of 0 to 0xffffffffffffffff.

"Unsupported sampler name"

• Explanation: The specified sampler name is not recognized.
• Solution: Verify that the sampler name is correctly spelled and supported by the node.

"Sequence loop count out of range"

• Explanation: The sequence loop count is outside the acceptable range.
• Solution: Ensure the sequence loop count is within the range of 1 to 1024.

"Invalid latent interpolation mode"

• Explanation: The specified latent interpolation mode is not recognized.
• Solution: Verify that the latent interpolation mode is correctly spelled and supported by the node.

"Noise strength out of range"

• Explanation: The noise strength value is outside the acceptable range.
• Solution: Ensure the noise strength value is within the range of 0.0 to 1.0.