Hey there, so I'll be releasing this one after testing it for a bit, I'm baffled at the high quality I've gotten so far. This LoRA was specifically trained on Illustrious v2.0 (the stable version) and I don't know if it works with other Illustrious models out there, so I'll have to test that aswell. I don't know if I even need to rework the dataset or not, the quality is amazing already, so I'll be waiting for feedback on this one... hopefully you guys enjoy it aswell. If you want to support me, you can either donate, like or follow for more updates, either one helps! Thank you!
Recommended settings:
Strength 0.9, lower for more flexibility. Same goes for your cfg, anything below 4.5 is perfect or recommended. Also try not to use too many LoRAs together as it can deminish the quality of your image fairly quickly, or a high cfg will also do that.
Thank you again for using my LoRA, you can support me through either donating Buzz, or you give a thumbs up, follow me or you can also use one of the links provided beneath my profile, anything helps really.
Description
Experimental version.
FAQ
Comments (21)
V1 better, sorry. V1 does magic.
V2 less creative, but can fix some complex compositions.
Here you may see diff https://civitai.com/posts/18228569
Yeah, v1 I had way more dark fantasy images on the dataset, v2 I added more variety, less dark fantasy/horror and more aesthetic/anime-ish gothic themes, and thanks for the feedback! I'll take another jab at the dataset, try and find a common ground for both styles to flourish.
@Comte I don't know what presets you used, but the first version really does magic, it changes the composition of the picture so amazingly that it's just great, for now the first version is forever in my heart, with it you get incredibly beautiful generations in my favorite style, thank you for that)
V2 is absolutely stunning, but less flexible than V1.
Yeah :/ I'll try and rework the dataset in the future, see if I can make it a bit more flexible.
Awesome sauce!
I really wish authors would put more effort into simply describing what their Lora even does instead of simply showcasing imgs.
The img generations look interesting but I have no idea what this Lora is even meant for. It's a shame rly, you miss out on greater exposure by being vague in description
It's a style lora, it's suppose to give you a style... That's very straightforward, idk what you're going on about.
ooooh, clueless
how is this getting 20 thumbs up? peanut gallery...
@deitychaser i imagine its because my expressed frustration is a sentiment shared by many others. Its not unique to this creator, i just happened to get annoyed enough at this one to say something, but its clearly highlighted something. Creators would be wise to acknowledge this.
@syndon Pictures say more than 1000 words but if you want and artcritique-esque description of the style you can run the example image through joycaption or gemini or chatgbt and they type it out for you very elaborated.
@deitychaser no need for hyperbole, a single descriptive sentence will do, and people will be more likely to use your model. It's a super low effort addition that could potentially increase your reach, why not do it?
@syndon Ah yeah, because you actually care so much for the exposure of something you apparently doesn't even understand is doing ("The img generations look interesting but I have no idea what this Lora is even meant for"). You're just complaining for the sake of complaining.
Over 2k downloads, over 11k generations on civitai since August. Exposure seems great, most people understand that a style lora is doing a style.. What style? Look at the gallery pictures lol
Look, the point is fair. Perhaps it could've been expressed in a more positive way to avoid the negative responses. But a simple sentence of "adds dramatic lighting and increases detail" or "gives artwork a style influenced by x" would be very helpful on a lot of these style lora's.
I've seen @syndon post multiple comments to this effect, and I believe they are trying to let creators know that - hey, if it was a bit clearer what this did and we didn't have to experiment to find out, you might get a lot more users. They mean well, despite the obvious frustration in the delivery.
If it were me though, I'd just ask the creator what the lora does specifically, that way the comment will stand as a resource for other users wondering the same, and not the start of a flame war.
But syndon's comment "you miss out on greater exposure by being vague in description" is facts. Some of us love to experiment, and will try dozens of loras. Others just want cool shadows and will use the first lora they find with a cool cover image that says it makes your shadows look better. Creators would be better served to explain what their lora's do... but that's assuming they actually care about getting users in the first place.
But if not, who cares. We shouldn't be criticising creators for how they publish their free resources that we do not pay for and are in no way obligated to use. Creators are the heart and blood of this community and they owe us nothing. Nor should we light up users for trying to help creators reach an audience. Provided their intention is good.
Maybe everyone just needs to watch their tone. Hahahaha love you all, I'm not your Dad, do what you want.
@Fuuji_9 I wasn't trying to be a douche with my reply, I sincerely didn't understand why this question came up in the first place, the tags I've added to the LoRA (you can see them right under the LoRA's name), describes exactly what this LoRA does with simple words, and it's a style LoRA, ultimately it gives you the style that I'm showcasing in the images. And since the day this question came up, I've already taken it as a review and have been adding more context to the LoRA's I've created after this one, but it just confused me, and since english isn't my native language, I might've come off like a douche, which wasn't the intention at all.
@Comte Hey man, I apologize for the tone of my original msg, it wasn't at all meant to be antagonistic. The thing with showcasing img with no context is you never know what aspect of the image is being influenced specifically by the lora (and not the Checkpoint the poster is using), so a simple description goes a huge way to identify exactly what the Lora is doing. Thanks for your efforts in contributing to our collective :)
@Fuuji_9 Thank you for eloquently explaining the true sentiment of the msg, You really put effort into that piece of writing. It was so articulate that I actually thought it was a Chat-gtp response for a moment xD
I'm thankful for the reminder to be more kind when communicating online, it's so commonplace to take up a combative stance because the internet has become so hostile and all-around anti-empathic. Thanks for being a light in the dark Fuuji
@Comte - I didn't think you were a douche at all <3
I appreciate y'all. <3
the comment makes sense, but you are assuming the author knows what the lora does enough to describe it in detail. I feel like a lot of style loras are so similar no one really knows what it does or how it is different.
Details
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.