A beginner's guide to the Fooocus-Api model by Mrhan1993 on Replicate

#coding #ai #machinelearning #programming

This is a simplified guide to an AI model called Fooocus-Api maintained by Mrhan1993. If you like these kinds of analysis, you should join AImodels.fyi or follow us on Twitter.

Model overview

fooocus-api is a third-party model created by mrhan1993 that provides a FastAPI-powered API for the Fooocus image generation software. Fooocus is a rethinking of Stable Diffusion and Midjourney's designs, learning from both to create an offline, open-source, and free image generation tool that automates quality improvements. The fooocus-api model allows users to access Fooocus's capabilities through a RESTful API, making it easier to integrate into various projects and applications.

Model inputs and outputs

The fooocus-api model accepts a wide range of inputs to customize the image generation process, including prompts, image seeds, guidance scales, and various control net parameters for image prompting and mixing. The model outputs the generated image paths and the corresponding seeds.

Inputs

prompt: The text prompt used to generate the image.
cn_img1, cn_img2, cn_img3, cn_img4: Input images for image prompting.
cn_stop1, cn_stop2, cn_stop3, cn_stop4: Stop values for image prompting.
cn_type1, cn_type2, cn_type3, cn_type4: Control net types for image prompting.
cn_weight1, cn_weight2, cn_weight3, cn_weight4: Weights for image prompting.
sharpness: The sharpness of the generated image.
image_seed: The seed used to generate the image.
uov_method: The upscale, outpaint, or variation method to apply to the image.
image_number: The number of images to generate.
guidance_scale: The guidance scale used in the image generation process.
refiner_switch: The refiner switch value for image generation.
negative_prompt: The negative prompt used to generate the image.
uov_input_image: The input image for upscale, outpaint, or variation.
style_selections: The Fooocus styles to apply to the image generation.
loras_custom_urls: Custom LoRAs URLs to use in the image generation.
use_default_loras: Whether to use the default LoRAs.
inpaint_input_mask: The input mask for inpainting.
inpaint_input_image: The input image for inpainting or outpainting.
outpaint_selections: The outpaint expansion selections.
outpaint_distance_top, outpaint_distance_left, outpaint_distance_right, outpaint_distance_bottom: The outpaint expansion distances.
performance_selection: The performance selection for image generation.
aspect_ratios_selection: The aspect ratio selection for the generated image.
inpaint_additional_prompt: Additional prompt for inpainting.