Home Creators Posts Import Register Favorites Logout

Content

Hey everyone! I've created a 1-click installer for the newest "ULTIMATE KONTEXT LORA TRAINING Ai-TOOLKIT WEBUI". Simply one of THE BEST Ai model training Webui I have ever seen! Made specifically for complete beginners! And you can train any LoRA for any model you want, including now for Kontext! … It's just amazing!

You can check out the video right here: https://youtu.be/s9eXc8PmOos

I also created a 1-click installer for the Webui for my Patreon supporters for Local AND Runpod, that you can find right here: https://www.patreon.com/posts/1-click-install-133520964

Here the link to the manual install guide: https://pastebin.com/P32MwBJG

Down below you will also find the link for the gigachad kontext lora for free for you to use!Download the LoRA:https://huggingface.co/Aitrepreneur/test/resolve/main/gigachad_k_000002250.safetensors?download=true

As always, supporting me on Patreon allows me to keep creating helpful resources like this for the community. Thank you for your support - now go have some fun😉!

Files

Comments

Maxed Out

As always, your manual instructions serve no purpose other than to overwhelm beginners into paying for your 1-click installer. Except this time you didn't even bother to make it for ai toolkit. 💀

Aitrepreneur

Huh?? What do you mean?? ALL the manual commands and instructions are there...what more do you want..?!

Maxed Out

The paste bin link takes me to "Manual Comfyui Kontext Models Install". This is not manual instructions for ai toolkit webui. Speaking of your ComfyUI instructions, paste bin? Really? Manually pip installing custom nodes like the manager cant do that... Oh wait you don't tell people that until they pay. I found your Flux v2 workflow as a beginner. Your manual instructions were so horrible I thought I could only use the workflow if I used your one click installer. I literally would have thanked you. Now when I look at them they are so ridiculously complicated that you obviously don't care if anyone can actually follow them. Like I said, your manual guide serves no other purpose except to overwhelm beginners to pay for your one click installer. You create the problem then sell the solution.

Aitrepreneur

Ok you are right about the manual instruction link, I did put the correct link in the video but not here. But tt's a mistake though, it happens. Secondly, THE REASON WHY it's preferable to install custom nodes manually is: 1) The manager has been broken for months and it often can't find the correct custom nodes to install, so the ONLY WAY is to install them manually 2) The manager works with an index and if your custom node is not correctly indexed or updated, the install through the manager will not work or will install a previous version of the custom node, so if you want to have the latest version of the node YOU NEED to install it manually, capisce?! I try to make everything as easy to follow as possible (I might not succeed but I try my best, I work long hours even with my current bad health to provide the best content possible, not everything is perfect but no one is perfect either), I give each one of my workflows FOR FREE alongside the explanation and commands on how to do everything manually AND YET you still somehow find a way to complain? Really?? Damn, that's a masterclass right there man...

Aitrepreneur

LOL WAIT! I just clicked on your profil! 🤣 Damn you got some nerves.... You are doing the exact same thing....I think that's the definition of hypocrisy right there 🤣🤣

Maxed Out

It's Max again. I don't have another account so if you block this one too you can rest assured I'm done lol. I actually agree with you about the manager thing. Seems like its always is having problems. You know I actually think I misinterpreted everything. When I saw your manual guide wasn't even for ai toolkit I was convinced you were a scammer lol. But I see you updated it and it's perfectly fine. 👌 Now I'm thinking I just saw everything in the wrong light and maybe some jealously over your paid member count blinded me alittle lol. Respect bro. You been at it for a long time. I do want to say that yes I am doing similar to you in that I also have a focus on Comfyui but many things are different. Not that everything is better but at least many things are different. Although I do see how from your perspective I was hating on you so hard then you realized I make ComfyUI workflows too. 💀 I won't lay it all out and linking to a post probably comes off like I'm trying to snag members from you. But if you ever are curious why I was going so hard at you about your guides, check my "🛠️ Manual Download" post for Flux. I get now why you do pastebin especially for ai toolkit but its just if you ever wanna see where I was coming from. Lastly, I actually think I should thank you. If not for you I would have never had the idea to make a Patreon focusing on ComfyUI. I'm no where near your giant number of paid members but it's promising. So thank you and I apologize for my overly harsh and unfair comments. I see the grind now. Respect. 🫡

Aitrepreneur

I unblocked you since you apologized but that really wasn't a nice thing to say. I already have enough problems irl as it is that I need to deal with this bs but thanks I appreciate the kind message

kocy

dude, thanks for the manual install process. :) I guess training loras really benefits from a 5090. Currently running a 4090 but running a runpod seems worth it. Thanks also for keeping it in free tier. I appreciate it. I reckon using a lora to decensor "stuff" would be an interesting field, too <.<.

MassMediated

Having an issue in AI-Toolkit where tagging my dataset images won't save/update. Tried one at a time, tried to create a new dataset folder entirely, did all the common sense solves. Have you encountered this/have any solves? Please and thank you!

Aitrepreneur

I'm guessing you are training on runpod, it's just a bug, don't go back to the dataset folder once you are done the caption, otherwise are you said it will delete the captions but the captions are still there once you do it once