Stable Diffusion is a large-scale Text-to-Image AI model. You can simply type prompts to generate images.
A prompt is a word, a phrase, or a sentence up to 500 characters long that describes what you want to generate in an image, including any details.
Negative prompts are the opposite of a prompt, allowing the user to tell the model what not to generate.
Quality refers to how many steps the AI makes over the image and that usually results in better quality (but not always)
Guidance is also called the CFG scale and it refers to how close the result matches your prompt. (It's the level of freedom for AI to generate art. So, it won't necessarily generate higher quality at max)
Image-to-image is transforming one image into a new one by combining it with a prompt, keeping its original style or characteristics. On Mage, you can simply insert an image and type a prompt to get a new image.