The fofr/sdxl-emoji tool, an AI model fine-tuned using Apple Emojis, is accessible via the Replicate platform’s API.
It enables image generation and manipulation based on input prompts and parameters. Users can input a TOK emoji of a man and provide a negative prompt for further customization.
Additionally, users can upload images for model use or provide a mask to define specific areas. They can specify output image dimensions and quantity. The tool offers solvers like Multistep, Heun, Discrete Karras, and others for refining images.
Users can adjust guidance scale, prompt strength, random seed, and refine style. They can control noise levels and refining steps and apply watermarks to generated images.
The model processes output in approximately 15.18 seconds, downloadable or viewable on the platform. It operates on Nvidia A40 (Large) GPU hardware via Replicate.
