Recent content by yano2mch

I've been experimenting with putting a book/game into the metadata of an image. It doesn't like tags it doesn't recognize, but it works with ones it's familiar with. Aside from very small books, or say shoving NES ROMS into a cover image so you get the game with the image, it doesn't seem...

Believe it or not, the prompts are within the image as a comment (including the hash of the models, the seed, loras, etc); However it's not like it's a secret. Had some of the whipped images for #3, and hadn't gotten #4 done yet.

Well here's a few of the pornmaster's results, they do look photo realistic.

I told it photo realistic originally; i can go for another model which is more likely to do it but that's kinda a tossup based on how it decides to interpret it. Maybe the pornmaster would give a real realistic rendition.

Well ran a book through and extracted a profile and data for Fran for @Haremfan, discussion commented here. Well the four scenes suggested (probably from the book) were: Chapter 1 Scene 1 - a visual example of fran (bondage commented on but not in the actual prompt?) Scene 2 - sexual encounter...

Mhmm. I'll make a note to revisit this later. Doing a single person nude or otherwise isn't hard. Doing multiple people I'm not experienced with. And sexual scenes won't work right unless there's a good Lora that helps push it in the right way, or there's a reference image it uses to rebuild via...

Could attempt, got some much better models than i had before for image generation. But as i haven't read any of the books I'd have to find the characters descriptions and/or interesting scenes of note to try and generate from. Also, often only 1 in 20 images are good to go right away. If you...

I tried a reverse lookup but couldn't find it. Yeah it's off topic, but oh well. --- Back to models. I am seriously impressed, the Gemma4 model in post #45 is really good at picking up cues, context, hints, replying naturally. Even smaller more subtle hints seem to be picked up on.

Trying out AI/Agentic Agents. Created a story-writing one to help. While i got it to write in a way that doesn't look robotic, the output tends to be as with a lot of things, more outlines and less a final product requiring three or four iterations and expansions before a said chapter is ready...

Here's results of the promising ones. edit: And putting photo realistic... Those actually turned out pretty darn well. Might revisit the previous requests to make photo realistic. It's been a long way since i was using SD1.5

The outline of what's there may be usable via img2img, but might even have to blur and shrink it and then upscale it again with how choppy it is. I'll try, probably with unholy desires as that model worked really well.

I'd only ask if you needed or had it use the other two books for sources. Personally I'd have it summarize each chapter, and get character profiles for each important character, then include that for reference, and if it mentions a specific scene or events the summary wasn't enough for, give it...

A note, the model is fast enough and very good i was trying for a higher quality Q6_K version... except... i keep getting issues. Mainly heavy repeat loops and getting stuck, use of 'wayward' as a descriptive filler text but used in places it didn't belong, trying to phonetically add sounds or...

Decided to try the multimodal features of a model. These are models that are made with text generation and visual information also is trained, so it is aware of both. So i gave it an image i think was from Pintrest, a funny cat Fluffy Von Whiskers. Then i asked the model what it saw. Here's...

For scenarios there's plenty of sources. Right now i tend to go with TelegAI (offshoot of characer hub) or Character Tavern, but StoryChan and JannyAI is okay too if you don't mind tags not being integrated correctly in the cards by default. And just cause they may not be labeled it, some of...