This DORA is designed to recreate the style of the artist Koe(Mixpi). Read the usage notes for details for each version. A Hires fix pass is recommended for all versions for best results.
All sample images were created directly out of ReForge UI with a Hires Fix pass. No inpainting or ADetailer was used.
Now that there is support for DORAs on most generation platforms, all future releases will be in the DORA format. These can be used exactly the same way as a LORA.
Please support the original artist wherever possible so they can continue to produce their art. Without them, LORAs such as these would not be possible.
This model is also available on Hugging Face and for generation on CivitAI.
USAGE NOTES:
For both versions, it is recommended to use full 1.0 strength. There are special tags for "koe-inaho" and "koe-mugi" to trigger this artist's most common characters. This isn't a character LORA, so results may be mixed when trying to trigger a specific character.
This artist likes to put lots of bandaids and bandages on their characters, so you may need to put "bandaid" or "bandage" in the negative if they are appearing unprompted. It is recommended to also add "depth of field, chromatic aberration" to prompts to achieve the artist's newer look.
PONY:
When using Pony, no trigger words are needed. V1 and 2 of this LORA was trained on Autism Mix. If you are using a different checkpoint, you may need go higher than 1.0 strength to maintain the style.
Recommended positive prompt:
score_9, score_8_up, score_7_up, score_6_up, source_anime
Recommended negative prompt: (you can adjust "hands" and "fingers" as needed)
source_pony, (hands:1.1)
NoobAI:
When using the NoobAI version, you should use the trigger word "koemixpi-style"
Recommended postive prompt:
masterpiece, best quality, koemixpi-style
Recommended negative prompt: (you can adjust "hands" and "fingers" as needed)
worst quality, bad quality, manga, screenshot, sketch, jpeg artifacts, (hands:1.1)
TAGGING NOTES:
The dataset contains over 250 images that were scored and ranked. The Pony dataset was also given the "source_anime" tag.
Scores/qualities were ranked in the order of most detailed/newest to least detailed/oldest.
Please do not violate the content policy