bing image creator prompt: “imagine garfield as an eldritch god, detailed oil painting, horrifying, a being of of terrifying, unknowable power”
resulting image was passed through Stablediffusion to be upscaled.
stablediffusion img2img parameters:
Prompt: lora:bodhor, lora:add_detail, disgusting body horror creature bodhor, oil painting
Steps: 30, Sampler: Euler a, CFG scale: 7, Size: 6144x6144, Model: dreamshaper_8, Denoising strength: 0.3, Clip skip: 2, Ultimate SD upscale upscaler: 4x-UltraSharp, Ultimate SD upscale tile_width: 512, Ultimate SD upscale tile_height: 512, Ultimate SD upscale mask_blur: 8, Ultimate SD upscale padding: 32, ControlNet 0: “Module: tile_resample, Model: control_v11f1e_sd15_tile [a371b31b], Weight: 1, Resize Mode: Crop and Resize, Processor Res: 576, Threshold A: 0.5, Threshold B: 0.5, Guidance Start: 0, Guidance End: 1, Pixel Perfect: True, Control Mode: Balanced, Hr Option: Both”, Mask blur: 8, Inpaint area: Only masked, Masked area padding: 32,
Resulting image from img2img had a face in the left eye, which i edited out
So that’s the real Flying Spaghetti Monster!
HP Lovecraft’s Garfield
Did the LORAs in IMG2IMG change the output design, or do they just help somehow with resizing?
They change the output. specifically wanted the loras to change the output design so it looks more like a mass of creatures fused together like the thing. For reference, here is the original output from bing:
And here is the bodhor lora on civitai:
edit: I linked the wrong lora before. here is the correct one, I think. I downloaded it some time ago. https://civitai.com/models/51288/body-horror-creatures
Garfield is not that cool.
