DoRA version added should be the better of the two versions make sure you download the correct version, added "draph" as a known token, and added "draph horns" to prevent some bleeding, so these two will work in both positives and negatives.
Remember to only use score_9, score_7_up, in positives, avoid 8 and other scores if you want the style to be at its strongest, applies to past and future loras.
LoRA made as a commission this one aims to get as much detail related to Asanagi's style, focuses a lot on getting more details in the eyes, and trying to get as close as possible to the faceshape, you might want to put "fang, fangs" in negatives if you don't want them, should work on all checkpoints based on PonyDiffusion V6, as always please do play with the weights!
Also I made a public spreadsheet so anyone knows what's one the pipeline, link
Any and all reviews are very much appreciated so I can know what to improve for this LoRA or for other models, thank you~
I have a Discord server where you can go and bother me about which style I'm working rn, or suggest ones to try my hand at.
Description
FAQ
Comments (8)
Thanks for your sharing! This is a very great lora. But I still have some questions...
Why can't I generate the same image as you even though I use the same checkpoint, prompt, seed, sampler, lora, and enable adetailer? Is it because I don't have "score_9 score_9_up" and other related files? The pictures I generated are placed in the gallery. I hope to get your answer, thank you!
When making this, did you try it at all with other LoRAs? It really doesn't like to be combined with Character LoRAs. Results in extremely messy faces, low-quality eyes, blobs for hands, etc.
Just wanted to say that this is a great lora. The best Asanagi style lora I found. It just works without needing to make that many changes in prompts.
Would be great if you keep expanding on it.
Thank you.
Base Model SD 1.5?
is this intented?
Another Asanagi style for my collection... yes... excellent...
New PDXL user here
How do you use and apply training data that is in downloads here?
I will tell ypu, your LORA is just goddamn flexible, i will show you. Thank you for your time and dedication
Of all the asanagi styles I've tried this is the best but I'm using Illustrious and it's really painful without this lora, can you make a version of Illustrious๐·°(৹˃̵﹏˂̵৹)
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.