วีดีโอ

I guess the question is do you have Python installed? I think I need to update it because there were some changes I made to the scripts that I applied to my online prompt generator where I use it there. I should be able to get it updated tomorrow to wear it automatically checks and installs necessary components to run. But you need to install Python and then there's a component called that needs to be installed

Best thing to do is just find an in-painting model that works best for you. There's quite a few out there.

Glad it was helpful!

You can get my whole styles.csv file on my Google share at share.xerophayze.com

@@AIchemywithXerophayze-jt1gg Thanks so much!!!

😂

Not yet! I think others have been doing it on my discord though.

I would recommend using stability matrix and then use that to install comfy UI.

Yeah sorry there was a couple videos that I didn't realize how low the audio levels were.

@@AIchemywithXerophayze-jt1gg No problemo, Could I please get your assistance with COmfy UI? Do you have discord server or could i dm you?

Lol, yeah someone else pointed that it a while ago 😅

That's great to hear. And just wait till you see the next updates. I'll be in a greeting my splitter functionality which allows you to split grids of images into multiple individual images or you can upscale or downscale existing images. I'll also be implementing a very simplified and easy to use version of in-painting that I think is going to blow everybody away.

😂 thanks

Yeah just in your prompt instead of specifying a background, specify a blank white background, or a blank black background or something along those lines.

I've had some people use IP adapter with this and get some pretty good results

@@AIchemywithXerophayze-jt1gg is this not do-able with SD1.5? I have been trying and i have every setting the same as yours but for diffuser there is not SD1.5 of it only xl

Yeah I don't know if SD 1.5 would be capable of doing this. I haven't used a 1.5 model for months

Glad you think so!

Done! It's on my Google share which you can find the link under my channel. It's under a folder called images

@@AIchemywithXerophayze-jt1gg You are the best mate, thank you ! And keep up the good work, wish you many more subs to come !

Lol, yeah I know. I'm working on it. 😅

Yeah I really hope the Dev starts working on it again.

@@AIchemywithXerophayze-jt1gg looks like its all about comfy now sadly

Extensions/config presets/

You know I was asked that on my last video on that and I'm not sure. I took a look around for some kind of software that just takes different angles of an object and I couldn't find anything specific.

When you click submit it generates the prompts. Below each prompt is a coffee button and a generate button. The generate button will generate an image based on the settings in the control panel.

Go ahead and join my discord, and you can ask me whatever you need to there. discord.com/invite/dCvWN7Nh

Thanks 😊 ​@@AIchemywithXerophayze-jt1gg

@@AIchemywithXerophayze-jt1gg Joined, thanks 🙏

Any time!

I'm not sure what you mean. I use the API back end on my web UI to service things like the image generation from my prompt generator. You can access the API from your web UI by scrolling all the way down to the bottom, and there is a link called API. That will open up the API reference for web UI

I tried a few things with that this morning. Got some pretty decent results, but there's a few things in the process I think I want to work out.

Welcome!

Yeah, I changed my mind on the splitter and made it available through GitHub. github.com/Xerophayze/splitter

Honestly some of the examples I was presenting there were not the best of examples. It actually works really well when doing things like rooms furniture, even people if it's set up right.

Yeah I was having problems generating a minotaur as well. Still diffusion has a lot of issues with horns sometimes

​@AIchemywithXerophayze-jt1gg I may have to get someone draw the character 😅

thank you we appreciate it ,,,i download zipfiles (splitter )from github and but its not working

You bet!

So you can do a couple things, in the stability Matrix data models folder, and maybe even under the stable diffusion subfolder you would create a new folder and Symlink it to your existing folder where you have models. or put all your models into the stability matrix model folder and then symlink the model folder under your other image gen package.

@@AIchemywithXerophayze-jt1gg Thanks for that advice. I'll try it.

Glad you liked it!

You inspired me. I did a whole other video today similar to this one and my other character sheet one, but this I go into detail about how to set it up and split the images so that you have an image set that you can use to train LoRa's. check it out later today: th-cam.com/video/A1IyMM9VO6A/w-d-xo.html

Your welcome

Your welcome

You inspired me. I did a whole other video today similar to this one and my other character sheet one, but this I go into detail about how to set it up and split the images so that you have an image set that you can use to train LoRa's. check it out later today: th-cam.com/video/A1IyMM9VO6A/w-d-xo.html

This was more for character generation and keeping the characters consistent. In a normal generation with an entire scene, that is an entirely different process honestly.

I just used my prompt generator to generate this prompt and it works great with Juggernaut. Wide view digital fantasy illustration, (Enormous Minotaur standing amidst ruins:1.2), Rule of thirds composition, Strong horns, Massive axe, (Ancient temple backdrop:1.3), Crumbling columns, Overgrown vegetation, (Eerie atmosphere:1.2), Moody lighting, Weathered textures, Dramatic shadows

​@AIchemywithXerophayze-jt1gg thanks but I am trying the minotaur in a character sheet

Ok make sure you watch my new video, it may help

Great to hear!

huggingface.co/thibaud/controlnet-openpose-sdxl-1.0/blob/main/OpenPoseXL2.safetensors

If i understand your questions, the best way to do this would be to use img2img with a low denoise to resize it. meaning, get your initial image, send it to img2img, set the image width and height to what you want, ie 1920x1080, set your denoise strength to something like .2 or .3 and generate. this will keep the same image but re-render it at the new resolution. this is essentially what hires fix is doing, but this way you have better control over the end size.

Yeah I know, but for some reason it doesn't work in this installation. Last night I reinstalled Forge, to fix another issue, so I'm hoping it'll fix this issue too.

@AIchemywithXerophayze-jt1gg Ah I see, yeah forge gets weird little hiccups like that. I agree that it's still way better, though.

That's interesting. I'll have to give that a try and see what the results look like. Thanks for the info.

I'll check it out!

My pleasure!

No problem 👍

You know I'm not sure. Most likely you can be need to drill into the folder structure to find out where the original files are and just replace them.