AI Wiki
One page per AI term: definition, how it works, an example and common misconceptions. Written from official documentation, with the verification date on every page.
Multimodal: vision language models (VLMs)
Image-text-to-text models, also known as VLMs, take an image input and can tackle various tasks.
Text-to-Image: Generating Images from Prompts
Text-to-image generates an image from a text description, also known as a prompt; a diffusion model takes the prompt and random initial noise.