Creation of images from texts

The Stable Diffusion “txt2img” text to image sampling script takes a text prompt as well as many option parameters for sample kinds, output picture dimensions, and seed values. In accordance with the model’s interpretation of the prompt, the script generates an image file. Users can recognize created photos as having come from Stable Diffusion by looking for an invisible digital watermark on them, but this watermark is rendered useless if the image is scaled or rotated.

Each txt2img generation will use a unique seed value that will have an impact on the final image. Users can use the same seed to get the same image output as a previously created output by randomising the seed, or they can use the same seed to explore other generated outputs.

produced image. The number of inference steps for the sampler can also be changed by the user; a larger value requires more time, whilst a smaller value could lead to visual flaws. The user can also control how closely the output image resembles the prompt by adjusting the classifier-free guidance scale value.[17] While use cases aiming for more specified outputs may choose to use a higher number, more exploratory use cases may choose to use a lower scale value.

Front-end Stable Diffusion solutions offer more text2img functionality by letting users change the weights assigned to different areas of the text prompt. By surrounding keywords in brackets, emphasis markers allow users to increase or decrease the emphasis on a keyword. Alternative strategy”Negative prompts” are sections of the prompt that are given less weight. Negative prompts allow the user to select suggestions that the model should avoid while creating images and are a feature featured in several front-end implementations, including Stability AI’s own DreamStudio cloud service. The specified prompts may be undesirable image features that would otherwise be present within image outputs due to the positive prompts provided by the user, or due to how the model was originally trained, with mangled human hands being a common example.