NEGATIVE REALISTIC / POSITIVE COMIC
This LoRA can either make a realistic scene into a comic style one, or make a comic image into a photorealistic one.
Works both in I2I and T2I.
Note: This is a concept LoRA, not a style LoRA. The model knows what a comic is, the LoRA just plays on these weights. The training dataset does not contain any comic images.
The sample weights range from 0 to 8 (full effect). The last sample is a comic image generated in T2I and made realistic in I2I with negative (-4) slider weight (empty prompt).
Hope you like it!
Description
FAQ
Comments (17)
what prompt do you use for going from comic to realistic with I2I.
Oh sorry I forgot to put it in the description, I don't use any :)
I can never remember which way is which. Negatives are more realistic?
It would be SO easy to just say "negative is realistic, positive is comic". But you don't say that. You buried it into the very last sentence where you say "[it was] made realistic in I2I with negative slider weight".
Just state it clearly at the top so we don't have to find it.
(👏)
It's free, and you're free not to use it. Sorry you had to "read the entire 4 sentences", I surely wasted your precious time.
@l226 Constructive criticism is hard to stomach, is it?
It's not that you wasted my time, it's that you went through all of the effort of creating a lora and then explained its use so poorly. Search "slider" and look at all the other ones out there - they clearly state what positive and negative do. Just take the advice, you don't have to get butthurt about it.
Yeah sorry I shouldn't have answered :)
@l226 I was frustrated and a little harsh but my intent was to help, however selfishly.
No worries, thanks for the feedback
Are you dumb or what? It’s an efffin slider! If it’s not clear, just run a test and the mystery solved!.
Next time, show us what you can develop instead of posting empty criticism or showing off your complete lack of knowledge.
@Crazy_Monkey Why waste my time testing when it could be so easily stated? That's the norm for sliders - the creator describes what it does if it's used via positive or negative strength. That's not an empty criticism, it's a clear, actionable, and easy change that makes it easier for users to interact with and use the lora. I was trying to help. What are you doing other than name-calling and trying to start some petty drama?
too many entitled n'wahs bashing on free loras, why does this happen?
@MilitAI Is there some sort of brownie point system that encourages you to simp for poor QoL or do you just prefer that loras be less user-friendly?
There is a standard for communication on this site that this lora, when I found it, fell short of - so I addressed it, and the creator updated their description. Notice how before my comment, there was another user questioning how to use the lora by strength? Notice how there haven't been any such comments since?
Also, I said nothing to bash the lora itself. I even posted an image to support it. jfc
Honest question: what exactly this LoRa does that Klein doesn't do with proper prompt? I mean, if you ask for "comic style, tracing", etc, it already produces excellent comic style.
It's a slider, it's not on/off it's able to handle effect's intensity.
Though I'd agree that for this one it's less useful.
If You check the sample with the redhead for example, the intensity varies, only with lora weight. Changing the prompt to modify intensity usually also modifies other things in the scene. If you have the perfect image with a seed and prompt, and just want to play on the "comic" intensity, sliders do that
combined it with body weight slider and Ani2Real loras, Anime to real convert process is kinda good. with Ani2Real lora you tell the model convert image, body weight slider to keep balance between head and body, with these you eliminate the cartoon/drawing weights. nice combo with this 3