Zankuro (Artist Style) [1.5/PonyXL]

0
0
0
0
AnimeAnime CharacterGirlCharacter
最近更新:首次發布:
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info
Anime,Anime Character,Girl,LoRA,SD 1.5Image info

Help fuel my passion! Even $1 makes a difference. https://ko-fi.com/citronlegacy

Check out all the Art styles I’ve made: https://civitai.com/user/CitronLegacy/models?tag=style

Check out my Pony Styles: https://civitai.com/user/CitronLegacy/models?tag=pony%20style


Please support the artist by checking out their page🔞: https://www.pixiv.net/en/users/7964

I really like the way Zankuro draws pointy ears.

Please experiment with the weight on this Lora.
0.8 gives a Zankuro experience but less weight like 0.5 or 0.6 produces nice anime styles and colors without the extreme Zankuro style.


Update Pony (03/18/2024) - Trained on Pony checkpoint with CivitAI's Lora Trainer.


Update v4 (02/17/2024) - New version is more style accurate and has more interesting poses.

🔞 Warning! 🔞 This Lora is very naughty because it matches Zankuro's style. Please use this Lora with caution.

Please experiment with the weight. Some images look nice with 1.0 weight while others look better with 0.7 weight.


Pony Version is trained with CivitAI's Lora Trainer.

Trained on: 881 images

Training Params:

unetLR: 0.0005

clipSkip: 1

loraType: lora

keepTokens: 0

networkDim: 32

numRepeats: 1

resolution: 1024

lrScheduler: cosine

minSnrGamma: 0

noiseOffset: 0.03

targetSteps: 2203

enableBucket: true

networkAlpha: 32

optimizerType: Prodigy

textEncoderLR: 0.00005

maxTrainEpochs: 10

shuffleCaption: true

trainBatchSize: 4

flipAugmentation: true

lrSchedulerNumCycles: 3


V4 Details

Trained on: 881 images

Training Model: AnyLoRA_noVae_fp16-pruned

flip_aug: False

Num of Repeats: 1

Unit is Epochs or Steps: Epochs

Number of Epochs or Steps: 10

Training Batch Size: 4

Total Steps: 2202

Resolution: 512

Network Dim: Dim 32 and Alpha 16

Lora Creation Process took: 00:43:21


Old v1 stuff

Suggested weight: 0.8 to 1.0

I recommend trying with weight 1 and if its too weird lower the weight.

Trained on: 27 images

Training Model: AnyLoRA_noVae_fp16-pruned

flip_aug: True

Num of Repeats: 10

Unit is Epochs or Steps: Epochs

Number of Epochs or Steps: 10

Training Batch Size: 2

Network Dim: 32

討論

最熱門
|
最新
發送
即將推出
下載
(0.00KB)
詳細資訊
類型
線上生成次數
0
下載次數
0
推薦參數
推薦搭配Checkpoint
推薦的重量
0.6
CFG
7
VAE

圖庫

最熱門
|
最新