Prompt Generators

Source Noise Variation: You can vary the source noise level to control the diversity of generated samples.

For example: "Generate images with low source noise."

Target Noise Variation: Adjusting the target noise level can impact the quality and realism of generated samples.

For example: "Generate images with high target noise for more abstract and stylized results."

Combination of Source and Target Noise: Combining both types of noise allows you to balance between diversity and quality in the generated outputs.

For example: "Generate images with low source noise and high target noise for a compromise between diversity and realism."

Noise Distribution Manipulation: You can specify the distribution of noise, such as Gaussian or uniform, to influence the style of generated samples.

For example: "Generate images with Gaussian source noise for smoother variations."

Noise Level Interpolation: You can request the model to interpolate between different noise levels to create a gradual transition in the generated outputs.

For example: "Generate a series of images with noise levels gradually increasing from low to high source noise."

Conditional Prompts: You can condition the generation on specific attributes or features by providing additional context in the prompt.

For example: "Generate images of cats with high source noise and low target noise."

Custom Noise Parameters: You can specify custom noise parameters, such as mean, standard deviation, or distribution type, to fine-tune the noise characteristics according to your preferences.

For example: "Generate [concept art] with [mean=0.5, std=0.2 Gaussian source noise]."