generated with stablediffusionXL. the title is a reference to a song by a certain russian band
Billy had a little bit too much but he's ok. He's always ok.
(generated with stablediffusionXL)
I did not want to accidentally send more comments to the person in the screenshots way by posting a link. It was a recipe for brownies. Here is the comment made by the author:
The recipe with the comments is easy enough to find online though.
I looked for the recipe this comment was made in, and in the comments, the original author of the recipe mentions replacing whole or part of the oil with applesauce which might explain why.
Prompt for the main image is:
gritty graphic novel art, a warrior holding up his hand which has turned into the head of a fire breathing dragon and shouting "forefathers one and all, bear witness!" speech bubble
Text is a lot easier on bing/dalle 3, for stable diffusion, its better to just do it manually, or use controlnet. There are loras like this one though: https://civitai.com/models/176555/harrlogos-xl-finally-custom-text-generation-in-sd
But its mostly for logos, and still takes a bit of prompting to get it to write what you want.
Before tv, phones and such, watching someone get sawed in half was probably the best form of entertainment you could have. Other than watching a play or something.
I asked microsoft copilot to interpret the lyrics to "Transdermal Celebration by Ween to create a prompt to generate an image with. The prompt it came up with was:
"A celebration of life and death, the turnover of life, and its continuation, depicted through the beauty of mutants and the transformative power of nature."
Self Destruct - kublai khan tx
yukikosan - midori (NSFW ALBUM COVER)
All songs that hit hard for different reasons... for me
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
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
parameters
Prompt: portrait of a female cyborg, shoulder, head portrait, bust, (human face:1.2), bloody cybernetics, biopunk, anime style, cyborg neck, cyborg shoulders, digital art, 80s anime style, inspired by (hr giger, zdzizlaw beksinski:0.7), horror, flesh and blood meets cybernetics
Negative prompt: robot, black and white, low quality,
Steps: 20, Sampler: Euler a, CFG scale: 4, Seed: 1256023903, Size: 1024x1024, Model: aamXLAnimeMix_v10 VAE: sdxl_vae.safetensors,
I wanted the look and feel of a photo that could have been taken on an old feature phone, so I added extra film grain after the image was generated.
Generated using realisticvision checkpoint.
I asked bing chat (microsoft copilot) to generate an oil painting inspired by the lyrics of "no children" by the mountain goats. I think the AI just used the prompt "an oil painting inspired by the lyrics of 'no children' by the mountain goats" instead of using the actual lyrics, and this was one of the images generated:
I thought it looked interesting, so I used Ultimate SD upscale in Automatic1111 with Dreamshaper checkpoint, upscaled 4x, with no prompt, 0.75 denoise, using the following extensions: tiled vae, tile/blur controlnet, and freeU,
I am now sitting on the exact same bench in the exact same position except two feet to the left
Bing prompt: surreal, colorful, psychedelic portrait of a man wearing a suit and tie, full body, desert, pastel colors, acrylic painting, neon, digital art, 70s art, weird west,
resulting image from bing was used in img2img in Automatic1111 scaled 2.15x using UnstableDiffusers YamerMix checkpoint with prompt: a man wearing a suit and tie in the desert, technicolor, digital art, psychedelic art, abstract art, surrealism, pastel colors, rainbow swirls, 70s art
ADetailer plugin was also used to generate swirly rainbow face with prompt: "abstract art, psychedelic art, 70s art, colorful rainbow swirls" at a high denoise
Generated with Bing Image Creator
Another image that uses both bing image creator and StableDiffusionXL
Original image created in Bing Image Creator with the prompt "photo of a man in a suit standing on a train platform in the desert, colorful, whimsical, bright pastel color scheme"
With almost the same prompt (i added sunglasses to the prompt) I used this as the input for img2img in automatic1111 with RealVisXL 2.0, cfg scale 5, scale by 2.13x, 0.6 denoise.
Bonus:
bing:
after img2img:
Generating this image was a two step process. The initial image was generated using Bing Image Creator with the following prompt: "space photography, nebula in the shape of the great wave off kanagawa", which produced the following image:
I wanted to use the image as a phone wallpaper, so I used that as the input image for img2img in Automatic1111 with RealVisXL Checkpoint with the following prompt: "a colorful wave of a blue and red nebula, space art, space photography, intricate and detailed,.outer space, nebula in the shape of waves, waves, ocean like". Scaled 2.13x, 0.5 denoise.
For reference, the original prompt that I used in bing image creator generates the following result in RealVisXL:
I could probably get something closer to the bing image with better prompts without using img2img, but I am lazy.
Prompt was something like movie poster for "Santa Saves Hanukkah".
Generated using Dall E 3 via Bing Image Creator
he looks kinda sad.
prompt: space photography, galaxy, stars, nebulas, (transparent ghost), lora:Aether_Ghost_v1.1_LoRA:1
SDXL Checkpoint: CopaxTimelessXL
Non upscaled version looks more ghostly, but I kind of like the subtlety of the upscaled version. Higher denoise values when upscaling makes it look more like a generic nebula.
This image was made using Automatic 1111.
Controlnet model is qrcode monster: https://huggingface.co/monster-labs/control_v1p_sd15_qrcode_monster
Settings used is: Pixel Perfect, Balanced, control weight: 1.2
The controlnet image was a photo of skull i found on google that I applied a black and white filter to.
