RULES/INFO - PLEASE READ FOR IMPORTANT INFORMATION

July 10, 2024, 05:49:07 am

First off, please register with the forum. No verification is needed.
If you don't sign in, you don't see any of the images in the messages...

Here are the rules spelled out:

1 - Absolutely no images that would break the law. No CP! You WILL be reported! Prompting teen nude will get you banned! 18+ ONLY! Tasteful Nudity ONLY!
That means no prompting anything degrading, ie on knees, mouth open. No sexual acts, sex toys, or masturbation. Only nude poses that are tasteful. This is not a porn site.
2 - Absolutely no racist or extremely offensive prompts or images.
3 - If you're not getting the result you're looking for, don't do the same prompt 10 times over, modify it or try a different model.
4 - Don't remove images you didn't create unless they break the above rules. Please make a note with the prompt info in chat.
5 - Do not queue more than 3 items yourself, be respectful of other people. Limit to 1 if the generation takes longer than a minute.
6 - Remove NUDITY or ANY NSFW PROMPTS from the Requested category immediately. Usually, the image will be detected and moved but if it doesn't,
you are REQUIRED to REMOVE it immediately. Nudity/NSFW content is only allowed in the NSFW category.

Failure to follow these rules will lead to a permanent ban. VPN's and unresolvable IP's are blocked again from generating due to scumbags.

I have started a campaign to upgrade the GPU's on the website. If you would like to help, please go to https://gofund.me/9f6915af.

If you are trying to add text to your image, then you need to use one of the Flux or Stable Diffusion 3.5 models. SDXL models cannot handle text. Select one of the models with [TEXT] in the title as these are using the higher fp16 encoder which provides a better result. Using more steps will give better results when adding text. Use the maximum steps for best results with text. Capitals and punctuation will affect the text outputed. Getting many words accurate can still be challenging. Recommend creating the image without text and then manually adding the text if there are alot of words.

If you are trying to generate nudity, use the SDXL model Lustify or Damn Pony or one of the LORA's like Fine Nude or Nudify. Original Flux and SD 3.5 are NOT trained for nudity. Flux Fill will not do nudity. FastFlux Unchained can do nudity. Only TASTEFUL nudity is allowed, this is not a porn site.

These models are image models, not chat bots. They have no intelligence! Just DESCRIBE what you want in the image. Better detail, better image. Don't repeat your description of any objects or it will create multiple objects. Describe each part completely and at once, then describe the next part of the image. Using more steps will generally give you a better image. Select a model with [NO TEXT] when not adding text in order to use the bigger and better fp16 models.

Read the Stable Diffusion 3.5 Prompt Guide: https://stability.ai/learning-hub/stable-diffusion-3-5-prompt-guide and Flux Prompt Guide: https://fluxaistudio.com/blog/prompt-guide to learn all aspects of prompting.

At the top of the screen just beside the title are the categories. If you are generating an image, after it has finished you can find the image itself and tools to modify the image in the Requested category (as well as a small copy on the generation page you can click to download the original .PNG image). Click the thumbnail image when under categories to view a larger version of the image. You can also use the Left/Right cursor keys to scroll through the images in full screen view. The larger view also has a 'Remove Background' button at the top as an option to edit the image.

Any images on this site may be used for free for personal use (screensavers, personal gallery, scrapbook art, social media posts (Not for profit), personal greetings, creative art (Not for profit), etc. If you're looking to profit from the images in any way, then please contact me at jdc4429@gmail.com to discuss terms.

Permission is also granted for free use of any images for company logo's or any work-related purpose where there is no direct profit from the image or images. For example, you can create an image for a presentation but not create an ad or a product you sell with the images. Like mug graphics or t-shirt designs.

Now that that is all out of the way, please enjoy the site responsibly and here are a few notes about some of the features of the site:

On the left you can find the menu to switch between the SDXL, Flux, and Stable Diffusion 3.5 models. SDXL is the fastest and offers 23 different models as well as 37 Loras giving you lots of variety. Flux has good prompt adherence. Stable Diffusion 3.5 also has pretty good prompt adherence and is better with realistic people.

One of the main reasons I started this site was a lack of places to see prompt info in order to learn how to prompt better. Each image also has a Download and Info button. The Info shows you which model was used, the resolution, the seed, the prompt, and various other settings that were used to generate the image. The Download button allows you to download the full-size image. The Remove button allows you to remove any bad images.

In the Top Right is a chat icon that when clicked opens up a Group Chat where people can communicate with each other. When someone sends a message a ding is heard and the icon shakes. You can also drag and drop MP3/OGG files into chat which go right to the music player as well as drag and drop JPG/PNG/WEBP image files into chat for people to view an image. If you hear 'Hey You Guys!' check the chat for important messages as it means I'm trying to get someone's attention who is probably about to be banned.

Any new updates to the website or tutorials can be found in the General Forum.
You can select 'Prompt Search' to search a database of 1.5 million prompts if you're looking for some new ideas.
You can select 'Image Search' to search the image descriptions of the images in the category folders. The NSFW and REQUESTED categories are not included in the search results.

By default, the search is set to match all words so all words must appear in the description to match an entry.
Don't try searching for 'a mage queen casting a spell' as the chances of a match are near impossible. Search for 'mage' or 'queen' or 'mage queen' or 'mage, queen'