# sdxl aiart stablediffusion aiartworks ai aiイラスト photography photographylovers photographysimulator fantasy gpt3 midjourney goodmorning novelai modelmix bloomingpanic bloomicoc bloomicyume stabledifusion urbanart aiartwork aiartcommunity maps landscape generativeart architecture rezero anime リゼロ エミリア mlpfim rarity

半日でSD15に対するモデル学習どこまでできるか検証
dreemboothで何回か"1girl,anime"を160～300枚程度で学習（re6まで）
finetuningで追加400～500枚程度学習(reE)まで
アニメ系のものだけ150枚抜き出してreEに対して作ったLoRAをマージ(00)
12時間でも形にはなる　使用GPUはA100
素材はniji/mid/NAI

7 36
2024-02-04

sdxl のイラスト finetuning こんな感じ〜

6 21
2023-07-10

Current state of my finetuning of
Please welcome AnyLoRA XL (with different substyles).
Using DreamShaper as refiner here

10 37
2023-07-06

Seems more like alignment than fine-tuning 🤔

Has anyone tried some sort of black box reward model for improving diffusion models?

My best finetuning runs, my loss function just goes sideways when training Stable Diffusion.

0 2
2023-04-02

Roburt and Nanobit! 🧡🤍

Two robuds ready to tackle the day (and possible you!). 😆

I wanna get around to drawing Livewire, Phobos and Malbolge too! Some other robot characters I mentioned but haven't yet shared since I am still finetuning some of their design elements. 💭

2 5
2023-03-30

I have a limited supply of PreOrders immediately available, and these will be shipped out within two weeks. There are 4 more pins coming, but they're much more complex, so they need just a bit more finetuning! Preorders from these will help make that happen sooner QwQ Thank you!!

6 19
2023-02-22

VQGAN‘s still my favourite. Recently been finetuning SD on VQGAN and am really happy with the results.

0 5
2023-02-04

gm fam 🙌

I also didn’t directly craft the prompt for this one!

0 1
2023-01-27

Trying to go for a classic-disney influenced anime style, pretty tricky without finetuning.

0 2
2023-01-15

Gods and Goddesses of War - Finetuning some settings for this style 🤗🤗 raw unedited

0 15
2022-12-16

Stable diffusion 2.0 is....interesting. structure is quite nice, still need to play with it more to see how prompting works best.

looking forward to finetuning most of all!

if you've played with it, give me your thoughts!

1 14
2022-11-26

I'm playing around finetuning some custom merges, and guessing how many ethnicities can I get with some exactitude.

1. Caucasian
2. Afroamerican
3. Asian (korean specifically)
4. Hindi

Feels some accurate for now... I will keep going!

1 18
2022-11-20

[#bloomingpanic

hewo i come bearing my mess of a bloomic oc, theyre an actor theyre going through stuff and most importantly they have two hands (ft my june design that im still finetuning arnd or wtvr teehee)

3 26
2022-11-13

finetuning HLA dreambooth model

2 31
2022-11-02

More examples from finetuning stable diffusion model with 10K satellite images via textual inversion, to merge the urban landscape with the micro-structure from slime mold.

0 8
2022-10-16

『氷の天使』エミリア - Emilia, Angel of Ice

My prompt could use a bit more finetuning, but I really love experimenting with the aesthetics of Emilia combined with themes of ice, snowflakes and feathers.

61 457
2022-10-14

massive thanks to for providing the compute used for finetuning! here's some more examples with the original image on the left, SD 1.4 VAE reconstruction in the middle, and the finetuned VAE on the right:

3 22
2022-10-10

FineTuningであるColab版dreambooth、何度かエラーが出て問題を調べようと実行したら最後まで走り切るという
某キャラクターを5枚、30分程度回してみた
さすがにこれだけだと精度的な問題はあるけど・・・

1 2
2022-09-29

Langsam wirds!
Morgen geht's ans finetuning (hoffe ich...!)
Danke für den tollen Stream heute Mittag, ihr macht mich so Happy! - Ich wünsche euch eine wunderschöne Nacht! ✨

4 75
2022-09-27