Community for AI image generation. Any models are allowed. Creativity is valuable! It is recommended to post the model used for reference, but not a rule.
No explicit violence, gore, or nudity.
This is not a NSFW community although exceptions are sometimes made. Any NSFW posts must be marked as NSFW and may be removed at any moderator's discretion. Any suggestive imagery may be removed at any time.
Refer to https://lemmynsfw.com/ for any NSFW imagery.
No misconduct: Harassment, Abuse or assault, Bullying, Illegal activity, Discrimination, Racism, Trolling, Bigotry.
AI Generated Videos are allowed under the same rules. Photosensitivity warning required for any flashing videos.
To embed images type:
“”
Follow all sh.itjust.works rules.
Community Challenge Past Entries
Related communities:
I've got prompts, and I'd love to see what y'all can do with them different than what I did! Also please enjoy Leonardo Elric, the fourth Elric brother
I’m a bit dense, could you explain this with an example?
You share a prompt and people try to create something similar to what the prompt describes?
I was thinking of trading prompts with people for either attempted improved accuracy OR to see their interpretation. It's not the most extensively though out plan yet
Oh I kind of assume that with all my posts :)
I love it when people take my prompt or concept and give it their own twist. But sure, I’m always happy to join in.
I'm always hesitant. I don't want to step on anyone's toes or offend
Yes, I know what you mean. People put a lot of work in their post (at least I like to think so :), and it feels like you're stealing their thunder if you reply with your own version.
Ok, anyway if you ever see one of my posts, feel free to play with the prompt and come up with your own interpretation. You're always welcome to do so :)
I'm in!
Aw yiss! Should we private message them, or how should we make this work?
Let's just post here, then others might join in!
I have never been able to get this one to work. It's a nightmare from my childhood. I'll post the one that's closest. Along with the prompt.
Visualize a scene where you are standing on a metal walkway grating, similar to those found in industrial settings, suspended above a factory floor. This walkway extends to the walls of the room, covering the entire area below. Instead of machines and workers typically seen on a factory floor, beneath the walkway are large cubic water tanks. These tanks are filled with murky water, and within them lurk shadowy, terrifying creatures. The perspective is from someone on the walkway, looking down at these tanks. The room is brightly lit by overhead lights, giving it a sterile, ominous atmosphere.
Imagine standing on a stainless steel walkway, perfectly centered and suspended at a mid-level height in a large, sterile white room. This room is brightly illuminated by harsh fluorescent lights from above. The walkway is a 3x3 metal grid, resembling a tic-tac-toe board, and is the only level present. Its metal grating allows for a clear view of what lies beneath. Directly below the walkway, large cubic water tanks are positioned, perfectly mirroring the grid above. These tanks are made of stainless steel and filled with murky water. Within the water, there are subtle, barely discernible shapes of ominous creatures, adding an element of mystery and foreboding. The walls of the room show signs of wear but are meticulously maintained. This viewpoint is from someone standing on one of the corners of the walkway, providing a comprehensive view of the walkway itself, the cubic water tanks below, and the imposing, sterile ambiance of the room
Wow, this was more challenging than expected :) Here’s my attempt in Midjourney:
Prompt: **aerial view of a man on a symmetrical grid of steel catwalks, overlooking large steel water tanks filled with monsters, in a white sterile laboratory with bright phosphorescent lighting, the room is built in a symmetrical design --ar 3:2 --v 6.0 --s 55
It took some experimenting to get the right view, symmetrical is definitely a keyword. And the man helps as a focus point to keep the proportions in order.
This is very close! It just needs to a square room, and the light must be bright white from above. Preferably murkier water. The AIs do seem to have trouble doing the perspective of someone standing on the catwalk.
Here's my attempt:
Prompt: Create a picture of a first person point of view: standing on a stainless steel walkway. Directly below the walkway are large cubic water tanks. These tanks are made of stainless steel and filled with murky water. Within the water, there are subtle, barely discernible shapes of ominous creatures, adding an element of mystery and foreboding. The walkway is a 3x3 metal grid, perfectly centrered and suspended at a mid-level height in a large, sterile white cubic room. This room is brightly illuminated by harsh fluorescent lights from above.
= = = =
I noticed that your prompt is a bit messy in regard to the features of the room you described: I found that it might be easier for the AI if you describe the things you want to see very separately and one after the other in a logical order, like you would paint them.
You wrote "stainless steel walkway, perfectly centered and suspended at a mid-level height in a large, sterile white room." Then you describe the room further, then you describe the walkway again, then the water tanks, and then again the room's walls.
I think it's better to focus on the walkway above the tanks first since they are the central objects, and then place them in the surrounding, the room. So focus points first, then the background. I'm not sure if this is understandable, but I hope it can help to clear up something about how the AI might struggle to understand your description.
My prompt is not very clear in this sense either, but maybe trying further in this direction helps.
I didn't manage to get it to create a cubic room either :-)
It's very good! And yes you are seeing two versions of the prompt. I have like 15 versions trying to get something accurate.
This is cool! Reminds me a bit like that old game Acid Factory
As for the prompt exchange, I was actually in the process of making a community for that kind of stuff. Never got around to promoting it, so it's been sitting unattended. You could use it / take it / change it if you wanted :) I'm happy to transfer it over to you to get going
What did you have in mind for it? Battles? My thought was if there were interest, we could make it a regular Contest type thing like we do with The weekly contest that we have now. I hadn't considered an entire other community, but it's an interesting idea.
Yep pretty much!
I was thinking that there could be a weekly pinned challenge, and on top of that people would be free to make their own challenges whenever. The name was inspired by the PhotoshopBattles subreddit since it's a similar format, (where there's some kind of theme or idea mentioned, and then the replies to the post are different attempts at creating it).
I was initially thinking of leaving it open to any kind of generation, text included, but image posts would likely be more popular. I haven't played around with generating poems or stories, and I don't know what the quality would be like.
I tried setting up a community since I thought it would be more organized. Another way could be to use a specific phrase and include it in the title
[CONTEST] Photo of a dreamscape with...
but it might be harder to find new contests/challenges with everything mixed in a big feed.
Regardless I'm happy to transfer that community, help promote it, or whichever format you'd prefer. I'd like more posts like this
I used Dall-E for these instead of SD since I can do iterations on it and have it focus on elements of the previous image easier.
None of the AIs are able to grasp the concept of a square room so far
While I work on this here is mine. Take something you did in life and push the visualization to its extreme.
Below is my example and what got me to take AI art seriously. I was trying to send a ridiculous thank you meme to a colleague as part of an inside joke and I couldn't find one that fit, so I went to Chat GPT and got started and asked it to push it even further into the extreme over several iterations. Below is what I came up with and neither of us like looking at it but we can't look away either!
