AnyText: Multilingual Visual Text Generation And Editing

• --model Sanster/AnyText
• GitHub: https://github.com/tyxsspa/AnyText (opens in a new tab)

AnyText has the capability to generate text on images that aligns with the style of the image and the prompt. It supports the generation of both Chinese and English characters.

Text to be written on the image should be enclosed in double quotation marks, for example: "DADDY"

Characters written in chalk on the blackboard that says "DADDY", best quality, extremely detailed,4k, HD, supper legible text, clear text edges, clear strokes, neat writing, no watermarks

Example negative prompt

low-res, bad anatomy, extra digit, fewer digits, cropped, worst quality, low quality, watermark, unreadable text, messy words, distorted text, disorganized writing, advertising picture

Example image