Instructions to use fidel1234/Krea2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use fidel1234/Krea2 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("krea/Krea-2-Raw", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("fidel1234/Krea2") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
Regarding the Selection Criteria(選定基準について)
I'll write in both English and Japanese.
I've noticed that the people you've chosen for your LoRAs come from a variety of backgrounds, including fashion models, gravure idols, singers, and actresses.
Do you have any specific criteria you follow when deciding who to make LoRAs of?
Are you also planning to create LoRAs of idols, voice actresses, news anchors, and so on?
英語と日本語両方で書いていきますね。
LoRA対象の人のチョイスがモデル、グラビアアイドル、歌手、女優と別れているようですが、
何かご自身で基準があって選ばれてる感じですか?
アイドル、声優、ニュースキャスターなども予定にありますか?
はじめまして
I trained mostly for gravure idol, since I love their works (especially baba-san).
It was started by just a exploring AI tech stuff, first LoRA train was on z-Image.
After switching to Krea and get good results of it, I decided to share the LoRAs
For the future, I am still focusing on training gravure idol, since I collect a lot of their photobooks, not sure about the other criteria
よろしくお願いします
ok! Thankyou for your reply!
返答ありがとうございます😊
Ths is is great content, thanks for creating and sharing.