May-be not fully in the theme but Guess what I was listening to wake-up and I kept exploring a bit more that song for this week challenge. Still has that Faeric tales, and it’s based on a song rather than on tale
Flux Dev (From getimg)
A faerie, with flamboyant punk-hair alternative style and cute butterfly wings, she wears a black leather suit and black leather boots, dance with a bearded dwarf wearing a red hat, full moon light, in London. high quality, 1970’s, urban fantasy, high quality
Miss some spikes, would be more efficient at rolling over kids and cyclists
Not sure whether it’s today’s theme, but let’s get a reptilian pope
The pope is a reptilian, at the church altar, holds a golden cross, conspiracy theory
Model being blue pencil XL
The covid was a biblical plague, don’t you know there is deciption of Modern ICU in medieval churches ?
A stained glass showing a doctor wearing a face mask helping an ICU patient plugged to monitoring and ventilator, grey’s anatomy style using Bing image generator
I won’t upvote you, so the post score matches the story. Here is a free hug though 🤗
couldn’t find your PSA though?
Thanks for your work, on making the weekly challenge happening.
I am a bit worried by logistic of having a thread for the “entries” and a thread the next week for the votes (Might be manageable with a script/bot) not sure whether you (or someone else) has the bandwidth to manage it.
for the rest, not sure whether we need to review the weight to limit the upvote weight compared to the extra points, or whether we shall accept a “first to posts wins” type of challenge, yesterday I try to add a PSA message, but don’t feel like it changed much the outcome
I think someone took that thread seriously
Here it a Google add I got
((Spiderman) wearing a pink) dress, on the podium at the fashion show, fashion magazine cover
Model : Pixel Ahuskay
Actually SD XL works pretty well compared to 1.5 Here are some results with a couple of models
Prompt is A Badass back tattoo, I generated 4 images (using comfyui) and have chosen one.
Art universe XL https://civitai.com/models/123313/art-universe?modelVersionId=622206
Blue Pencil XL https://civitai.com/models/119012/bluepencil-xl
Actually surprised that it looks more video-game than anime
Zavy Chroma XL https://civitai.com/models/119229/zavychromaxl
And using the Flux workflow from getimg.ai
Remember that tons of website are basically running the same code (usually stable diffusion) under the hood and just provide you their own UI. Then you have Dall-E (In Bing image generator) and Midjourney which have their own code under the hood.
Also, beside a bit of luck, you can define how much you want the model to follow your prompt and other similar stuff, let alone the whole prompting part.
I know, there is a kind of “boost” when the thread still pops in the “sort by new feed” . But on the other-hand, especially when challenge starts on the week-end (this one it’s a borderline case). Moreover, I also think about that game when I see the thread, not 3 days latter. Sure I could have kept the image, and post-it this morning rather than yesterday night, but feel weirds to do.
For that one, I went back to SD 1.5. using «Epic diffusion» and Euler ancestral samplers
One man in black suit, with a wedding ring, plays a beautiful piano, in prohibition style speakeasy bar. BREAK BREAK BREAK On the background, a detective with a long coat and a moustache enjoys a fine whisky, 1950’s film noir,
P.S Also, no fan of the first 24h penalty, especially when the challenge is posted on the week-end, but here is my contribution anyway, if not I’ll forget.
Let me introduce you : Superclown the hero that’ll make you laugh
A flying clown, wears a red cape and a blue jumpsuit, 1950’s superhero comics, flies over the city, SUPERCLOWN title
Model : Pixel Ahuskay https://civitai.com/models/537844/pixel-ahuskay
I was expecting some spoiler about Cixin’s 3 body cycle
Do you have a PC with a gaming GPU? If yes, you can run comfyui, automatic1111 or easy diffusion on your PC. It’s not that complicated (but require to be familiar with installing/configuring a software. Which considering lemmy’s audience is most likely the case)
If not there is tons of online services, some free (with restrictions), some with subscription, some with paying image credits. And you can check the faq of each services for details. Not all online SD services let you use lora, don’t ask me why
Actually, I tried to get a mecha first, but it didn’t got the cool vibe of the Shooter :)
For reference here is what I done forChallenge 5 a new sport can be fun to re-run-it one year later may-be I try in the week, but I think I’ll stay with the Turk shooter cyberboy for this challenge
Indeed, it’s just futuristic Olympics
Workflow
I used the cool Turkish guy from the Olympic as a control-net image https://sopuli.xyz/post/15435081 (Hard edge, 45%). Surprised that nobody has done-it yet
The prompt is A futuristic cybernetic boy with luminous blue hair, dressed in a iridescent white shirt, [flashy earrings and sick ass tattoos] has a chrome cybernetic arms, holds a sci-fi gun, in a stadium, sci-fi, cyberpunk, tournament
Model is Reproduction v3.31
My negative prompt is my very standard nsfw, Disfigured, blurry, extra fingers, extra limbs
Note that I struggled to get nice cyber gear, but once I added the sick ass tattoos, suddently it merged with chrome and neon and got that cool result
With some luck, I even got some olympics ring in the background, that’s neat
And for the fun, bing version, a bit more romantic