ComfyUI Node: KSampler Matrix Lab

KSampler Matrix Lab:

KSamplerMatrixLab is a powerful node designed to facilitate the exploration and experimentation of different sampling and scheduling combinations in AI art generation. This node allows you to systematically test various samplers and schedulers to determine the optimal configuration for your creative projects. By automating the process of running multiple combinations, KSamplerMatrixLab saves you time and effort, enabling you to focus on the artistic aspects of your work. The node is particularly beneficial for artists who want to fine-tune their models by experimenting with different parameters to achieve the desired visual effects. It ensures that you can explore a wide range of possibilities without manually adjusting each setting, thus enhancing your workflow efficiency and creative output.

KSampler Matrix Lab Input Parameters:

seed_mode

This parameter determines how the seed value is applied across different combinations. The options are same_seed_for_all, which uses the same seed for every combination, and increment_per_cell, which increments the seed for each combination. The default is same_seed_for_all. This setting impacts the randomness and variability of the generated outputs, allowing you to either maintain consistency or explore variations.

cell_scale

This parameter controls the scaling factor for each cell in the grid. It is a float value with a default of 1.0, a minimum of 0.05, and a maximum of 4.0, adjustable in steps of 0.05. Adjusting the cell scale can affect the size and detail of the generated images, providing flexibility in how you present and analyze the results.

font_size

This integer parameter sets the font size for text within the grid, with a default of 22, a minimum of 8, and a maximum of 96. It allows you to customize the readability of labels and annotations, ensuring that information is clear and accessible.

padding

Padding determines the space around each cell in the grid. It is an integer with a default value of 10, a minimum of 0, and a maximum of 96. Adjusting padding can help you manage the layout and spacing of the grid, making it easier to view and interpret the results.

header_height

This parameter sets the height of the header in the grid, with a default of 72, a minimum of 24, and a maximum of 256. The header height affects the amount of information displayed at the top of the grid, such as metadata and configuration details.

left_header_width

This integer parameter specifies the width of the left header, with a default of 180, a minimum of 48, and a maximum of 512. It allows you to adjust the space allocated for labels and annotations on the left side of the grid, enhancing the clarity of the displayed information.

background

This parameter allows you to choose the background color of the grid from options white, gray, and black, with the default being white. The background color can influence the visual contrast and overall aesthetic of the grid, making it easier to focus on the generated images.

show_grid_lines

A boolean parameter that determines whether grid lines are displayed. The default is True. Showing grid lines can help you visually separate and organize the different combinations, making it easier to compare results.

show_cell_labels

This boolean parameter controls whether sampler and scheduler labels are repeated inside every grid cell, with the default being True. Displaying cell labels can facilitate detailed inspection and comparison of individual combinations, especially when zooming in on specific areas of the grid.

show_run_header

A boolean parameter that decides whether a top header with model, VAE, CLIP, steps, CFG, and denoise metadata is added. The default is True. Including a run header provides context and reference for the configurations used in generating the results, aiding in analysis and documentation.

continue_on_error

This boolean parameter determines whether the process should continue if an error occurs during sampling. The default is True. Allowing the process to continue on error ensures that other combinations are still tested, even if some fail, providing a more comprehensive exploration of possibilities.

max_combinations

An integer parameter that sets a safety limit for the number of sampler and scheduler combinations, with a default value defined by DEFAULT_MAX_COMBINATIONS, a minimum of 1, and a maximum of 1000. This limit prevents excessive computational load and ensures that the process remains manageable and efficient.

KSampler Matrix Lab Output Parameters:

cells

The output parameter cells is a dictionary that contains the results of each sampler and scheduler combination. Each entry in the dictionary corresponds to a specific combination and includes the generated image and any associated errors. This output is crucial for analyzing the effectiveness of different configurations and selecting the best options for your artistic goals.

KSampler Matrix Lab Usage Tips:

• Experiment with different seed_mode settings to explore both consistent and varied outputs, depending on your creative needs.
• Adjust cell_scale and padding to optimize the layout and presentation of your grid, ensuring that results are clear and easy to interpret.
• Use show_cell_labels and show_run_header to provide context and clarity, especially when sharing results with others or documenting your process.

KSampler Matrix Lab Common Errors and Solutions:

KSampler Matrix Lab refused to run <total> combinations. Current max_combinations is <max_combinations>.

• Explanation: This error occurs when the number of sampler and scheduler combinations exceeds the set max_combinations limit.
• Solution: Increase the max_combinations parameter if your system can handle more combinations, or reduce the number of selected samplers and schedulers to fit within the current limit.

KSampler Matrix Lab failed for sampler '<sampler_name>' and scheduler '<scheduler>': <exc>

• Explanation: An error occurred during the sampling process for a specific sampler and scheduler combination.
• Solution: Check the specific error message for details, and consider adjusting the parameters or configurations that might be causing the issue. If continue_on_error is enabled, review the results for other combinations that were successfully processed.