I just added a new Wallpaper to the shop. This one is actually animated using an awesome free software named Lively Wallpaper.
Do not hesitate to write directly to me if you have a problem with the wallpaper.
2023-09-24 09:23:50 +0000 UTC View Post
Here are a few pictures (based on prompts from some of my most succesful art from CivitAI).
All the pictures have been created using a 0.5 mixture of Kohaku-XL and Counterfeit XL and a LoRA weight of 0.6.
The workflow is my usual, but starting at 680x1024 then img2img at 1024x1536 and finally img2img at 1080x192...
2023-09-24 08:54:09 +0000 UTC View PostAs I've anticipated in the previous post, here's the LoRA I've trained to experiment with SDXL. I've used it with Kohaku-XL and Counterfeit XL, but it's not nearly production ready (and that's the reason I'm releasing it only to subs).
I highly suggest using the standard SDXL VAE and a negative embedding if you want t...
2023-09-24 08:47:46 +0000 UTC View Post
Hi everyone!
Today, I've spent most of my day playing with some Stable Diffusion XL Anime models to see how they performed.
Here are a few things I've noticed:
So this one is kind of different from what I usually do. The first thing you'll probably notice is that this is not trained over AI generated images, but frames from the Aladdin movie (and that's one reason why this is released as a subscriber exclusive).
This was an interesting experiment, but I am not completely satisfied with the final result. I have another Disney based LoRA I'm going to release in the next few weeks.
As usual let me know what you think!
2023-09-22 08:57:55 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers.
2023-09-22 08:47:24 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
I highly recommend trying this one also with other LoRAs, I practically use this on every picture.
This one is a bit different, it's a slider, so increasing its weight will make the pictures more colorful, while using a negative weight will make the pictures grayscale.
I usually keep values between 0.5 and 0.8, but feel free to experiment also with 5.0 and -5.0 to see how it works.
Let me know what y...
2023-09-22 07:57:56 +0000 UTC View Post
Hi everyone!
I just wanted to share with you what I'm working on at the moment:
Hi everyone!
I've updated the workflow extension (in reality I've updated it 2 weeks ago, but I was lost in Starfield...).
The script now has an orientation parameter that guesses the picture orientation by looking at its size (but you can also specify it manually).
Have some ideas on how to improve the script? Let me know in the comments below.
P.S. New features are also coming for Ranbooru as soon as I clean the code...
2023-09-20 21:10:13 +0000 UTC View PostHi everyone!
Patreon just added the possibility to sell digital content on their platform while being integrated with the rest of their website. Don't worry, my LoRAs and scripts will remain free for all my subscribers as usual, but I'm going to add a few goodies on my shop for those who are interested.
Right now, I'm starting with a pack of 5 pictures I've worked on in the last few days. These have been highly edited in Krita and took a lot of work to reach this ...
2023-09-20 20:40:03 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers.
2023-09-18 07:28:16 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
You don't need a word to trigger the LoRA, but you can add "inzaniak" to the prompt to increase its effect.
Let me know what you think!
2023-09-18 07:26:55 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers.
2023-09-14 07:41:59 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
You don't need a word to trigger the LoRA, but you can add "inzaniak" to the prompt to increase its effect.
Let me know what you think!
2023-09-14 07:39:26 +0000 UTC View PostThis is a LoRA that I developed a few weeks ago, mixing Contact with other LoRAs that I don't remember. I wanted to release it publicly, but I was never 100% convinced so I decided to release it as an extra for my patrons!
I recommend using a weight of at least 0.8 as it is particularly "soft".
2023-09-12 09:07:11 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers.
2023-09-10 09:32:30 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
You don't need a word to trigger the LoRA, but you can add "inzaniak" to the prompt to increase its effect.
Let me know what you think!
2023-09-10 09:31:15 +0000 UTC View Post
Here we have the first post of "The Lab" series.
This is the LoRA I made a while ago to use as a base for my Emerald_v2 model. It has been trained over 65 pictures created using the Mistoon_Emerald_v1 checkpoint.
Let me know what you think!
2023-09-06 20:52:32 +0000 UTC View Post
Hi everyone!
Lately, I'm receiving a ton of support from my community, so I wanted to share some exclusive content with my Patrons.
So with today I'm starting a new series of posts named: "The Lab".
When I work to create a LoRA I usually go through multiple iterations before getting to a point I am fully satisfied. These intermediate steps are usually abandoned on my drive, but they can often give interesting results, so I'm going to share them with you.
In these posts...
2023-09-06 20:45:58 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers.
2023-09-06 07:26:12 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
You don't need a word to trigger the LoRA, but you can add "inzaniak" to the prompt to increase its effect.
Let me know what you think!
2023-09-06 07:24:47 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers.
2023-09-02 20:40:23 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
You don't need a word to trigger the LoRA, but you can add "inzaniak" to the prompt to increase its effect.
Let me know what you think!
2023-09-02 20:39:16 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers. (Made using Mistoon_Sapphire_v2)
2023-08-29 19:59:35 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
You don't need a word to trigger the LoRA, but you can add "inzaniak" to the prompt to increase its effect.
Let me know what you think!
2023-08-29 19:56:29 +0000 UTC View Post
Hello Everyone!
Yesterday I've discovered this repo on Github:
ostris/ai-toolkit: Various AI scripts. Mostly Stable Diffusion stuff. (github.com)
In this repo there is a script named LoRA Slider Trainer, which is based off the work by p1atdev/LECO and ...
2023-08-26 09:50:35 +0000 UTC View Post
Just a few pictures showing the capabilities of my latest LoRA available to subscribers.
2023-08-25 09:53:22 +0000 UTC View PostYou can now download my latest LoRA in Early Access.
You don't need a word to trigger the LoRA, but you can add "inzaniak" to the prompt to increase its effect.
Let me know what you think!
2023-08-25 09:51:14 +0000 UTC View Post
Hi everyone!
I've released on GitHub my latest script:
I've used this one for the last few months to help me streamline my workflow and create pictures faster. With this script you can quickly select a resolution, denoising strength and number of pictures with a single click.
If you want to learn the workflow I'm using with this script, you can check my medium article:
Hi everyone!
I wasn't fully satisfied with the current content for my amazing patrons, so I've spent the day working on a complete package of all the LoRAs I've released here and on CivitAI!
This package also contains a preview for each model created using the same exact prompt to help you navigate the huge amount of LoRAs!
Let me know if I've missed something (it took a lot of time to make, so I may have missed a LoRA or two 🤣).
You can find the download here:
<... 2023-08-23 21:14:09 +0000 UTC View Post