40th Lora, twentieth style.
(Noob/Illustrious) Trigger Word: artist:magnetus
Can innately do (some) characters they've drawn, see showcase for examples.
Support Magnetus:
https://magnetusart.carrd.co
Will remove if asked.
(All images are upscaled x2 using Hires. fix with 4x_foolhardy_Remacri as the upscaler)
Some images were refined with Noob Cyberfix
Description
FAQ
Comments (4)
could it be possible to update / remake the model? it is a bit weak and comes with plenty of error, the ones for pony gave better results but the ones from pony suffer from the lower quality results that the model inherently have. it would be nice to have a good magnetus model for illustrious that were stronger on the style with much less errors and better conformity to the base checkpoints and prompts.
my point is exemplified by all the art posted bellow this model that use it at low strength or merged with another model that it barely even comes out.
I intend to update most of the style Lora's I've made already to be re-trained on the Epsilon-pred 1.1-Version of Noob.
Most of the gen's below are primarily one user choosing to mix multiple Loras to get their results. I think this version works well on Illustrious Mixes, but regardless I can't control how someone chooses to generate images using this style besides pinning my gens to guide users.
i was not saying it was bad to mix it, its fine if people want to do so, but proportionally, sorry if it came out this way
Awesome style!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.