Good to see you! Try the new version!
If you would like to use my Checkpoint Online Generation, please click here. I added it because my model cannot be generated online with civitai.
LINK. TensorArt. yodayo. SeaArt.
And if you like it, please download it, write a review and press the Like icon. All of this will be my motivation.
各バージョンの日本語説明はAbout this versionにあります。
To prevent unauthorized reproduction and for the benefit of those who wish to build LoRAs, we will be publishing models on other sites in the future. However, commercial use is prohibited. (01/21/2025)
All rights reserved. 無断転載を禁止します。(02/4/2025)
V.20
This version update inherits the features of the previous version 1.0, and has been adjusted to enable the illustrations you want to express with short prompts.
The model is basically anime-style, but you can add PVC, 3D, figures, rear sticks, blenders, etc. to the prompt to give it a PVC taste or a three-dimensional finish.
The reason for this model is that I wanted to use multiple tastes in one because my PC is old and it takes a long time to switch checkpoints.
Also, I don't have the performance to do training, so I created this model by merging.
Base model of checkpoints used for merging in the past 6 months
white unicorn
Pony Diffusion V6 XL
SakuraSushi XL
Omega Pony XL anime style
3x3mixXLtypeE
copycat-PVC
DucHaiten-Pony-XL (no-score)
PD for Anime
AuthakuMix
coco-anime pony model
Recommended Prompt
Heavyweight
(score_9, score_8_up, score_8, score_9, ASCII score_9), (source_anime), (masterpiece best quality), (beautiful atmosphere), (8K, UHD), (sharp eyelid, sharp eyelashes, thick eyelashes, long eyelashes), (sharped face, beauty makeup), (beautiful skin, beautiful face), (curvaceous figure), ( perfect anatomy), (beautiful detailed face, beautiful oval face), (detailed embroidery), (sharp out line), (perfect hands), foreshortening, ( natural lighting, dramatic lighting,
Light weight
(score_9, score_8_up, ASCII score_9), (source_anime), (masterpiece best quality), (8K, UHD), (detailed embroidery), (sharp out line), (no nose), skin dentation, detailed backgrounds, cinematic scene,
Add PVC, 3D, rear sticks, and blenders to the * prompts as appropriate for the situation. Also add the same to the negative prompt. (Because depending on the LORA used in conjunction, it will be more cartoon-like or more realistic.)
If you are concerned about the nose of the illustration you have created, please try ((nose)).
SDVAE
sdxl_vae
CGF scale:7
Sampling step:25
Sampling method: DPM++ 2M Karras(or your choice.)
Thank you for reading this long.
Have fun!!
Description
This version update inherits the features of the previous version 1.0, and has been adjusted to enable the illustrations you want to express with short prompts.
The model is basically anime-style, but you can add PVC, 3D, figures, rear sticks, blenders, etc. to the prompt to give it a PVC taste or a three-dimensional finish.
The reason for this model is that I wanted to use multiple tastes in one because my PC is old and it takes a long time to switch checkpoints.
Also, I don't have the performance to do training, so I created this model by merging.
Base model of checkpoints used for merging in the past 6 months
white unicorn
Pony Diffusion V6 XL
SakuraSushi XL
Omega Pony XL anime style
3x3mixXLtypeE
copycat-PVC
DucHaiten-Pony-XL (no-score)
PD for Anime
AuthakuMix
coco-anime pony model
Recommended Prompt
Heavyweight
(score_9, score_8_up, score_8, score_9, ASCII score_9), (source_anime), (masterpiece best quality), (beautiful atmosphere), (8K, UHD), (sharp eyelid, sharp eyelashes, thick eyelashes, long eyelashes), (sharped face, beauty makeup), (beautiful skin, beautiful face), (curvaceous figure), ( perfect anatomy), (beautiful detailed face, beautiful oval face), (detailed embroidery), (sharp out line), (perfect hands), foreshortening, ( natural lighting, dramatic lighting,
Light weight
(score_9, score_8_up, ASCII score_9), (source_anime), (masterpiece best quality), (8K, UHD), (detailed embroidery), (sharp out line), (no nose), skin dentation, detailed backgrounds, cinematic scene,
Add PVC, 3D, rear sticks, and blenders to the * prompts as appropriate for the situation. Also add the same to the negative prompt. (Because depending on the LORA used in conjunction, it will be more cartoon-like or more realistic.)
If you are concerned about the nose of the illustration you have created, please try ((nose)).
SDVAE
sdxl_vae
CGF scale:7
Sampling step:25
Sampling method: DPM++ 2M Karras(or your choice.)
Thank you for reading this long.
Have fun!!
FAQ
Comments (8)
Great model, but I want to know why it turns the color image gray at the end.
It says uploaded 3, but I only see 2. Am I misunderstanding?
Here is the notice. Please read this information.
On December 5, my models will be archived or made private indefinitely and will not be available for download thereafter.
The reason is that I started AI illustration two years ago at this time. And I downloaded many LORA models through civitai. December 5 is the anniversary of my registration to civitai. I had intended to make it the day I started AI illustration, but I thought it was too sudden, so I decided to make it that day.
お知らせです。是非ご一読くださいませ。
12月5日に私のモデルは無期限にアーカイブまたは非公開とし、以後ダウンロードはできなくなります。
理由というか、私がAIイラストを初めたのが2年前の今の時期です。そしてcivitaiを通じて数多くのLORAモデルをダウンロードしました。つまりこれは私の少しもの恩返しでした。12月5日は私がcivitaiに登録した記念日です。本当はAIイラストを始めた日にしようと思ったのですが、余りにも急すぎると思いその日にしました。
This is a very good model! Actually my favorite. Sad to hear it will be gone soon. Thanks for creating this!
Announcement.
I really agonized over the whole thing and decided not to keep it private. The reason is that I felt the possibility of using NoobAI and Illustrious and decided to try a little harder.
アナウンス
本当に物凄く悩んだ結果非公開にするのを辞めました。理由はNoobAIやIllustriousを利用して可能性を感じてもう少し頑張ってみようと思いました。
great model, but in the end, it turns colored images into grayscale. I'd like to ask which VAE I need to download to restore it to normal. Please provide the download link for the VAE.
Let me know if you need further clarification