Dear All friends, Good news! more than 100 New Flux loras are released on Shakker.ai,please visit below website link and try them for fun. thank you for your long time support to Happy Ai Art Studio. Website link: https://www.shakker.ai/userpage/a841009572bf4eec80d159321a0cfd13/publish
❤️Hello all friends~ ~大家好~~みなさんこんにちは~~
안녕하세요 여러분~~Bonjour~~привет~~Hola~~नमस्ते~~مرحبًا
Welcome to Happy Ai Art Studio❤️
Introduction
I like ART and painting, so I try to explore more fun in this series. This lora model is good at creating landscape, natural scene, and human. It creates not only a more serene painting atmosphere but also provides greater flexibility and give something genuinely new and engaging to creators. I sincerely hope you like the LORA and enjoy using it.
Kindly note:
Dear all friends, if you want to create the same/similar as the images in the gallery, highly recommend to download this lora into your own computer. Because the gallery's images are with lora weight of 2-4, in contrast, Civitai's online creation is with 1-1.5 of normal lora weight, if over 1.5, then some images will become weird or ugly on Civitai 's generation. I am 100% sure the lora is all good as "tons of" images tests were done before its release. thank you for your attention and kind support to Happy Ai Art Studio.
Triggers: A. J. Casson style
(If without triggers, it still works but with limit painting effects. If the result is not towards the style you like, recommend to adjust the Lora weight or add some style words that you may like into Positive Prompts,for example: "painting,oil painting,peaceful,nature,landscape,artistic style,vivid color,...etc ."
And if needed, add some words into the Negative Prompts to avoid realistic style, like for instance :" realistic,photo,3D, 3d illustration....etc")
(Simply drag the images from my gallery into your computer, then you can see base model ,sizes and related settings)
Tips:
Tips:
About LoRA Weight:
For Creation On CIVITAI online setting:
recommended to start with 1, and then increase it step by step.
For Home Computer Users setting：
LoRA Weight: Between 2.3-3.5 range, recommended to start with 2.8
(if 2 or below 2, it may has little or no effects), If over 3 , some weird photos may appear)
(pls find below some comparison photos for different lora weights)
❤️For the optimal performance, recommend to download this lora model into your computer to generate images instead of creating it on Civitai only, because due to different settings bettwen Civitai Internet and home computer, the image may be different.
For example the original image with 1344 length will be shorten into 1216 length automatically in Civitai as below photo.
Base Models:
Although various base models can function, models with realistic styles like the SDXL 1.0 series base models, DreamShaper XL are preferred for optimal performance, but you can also try pairing it with other base models to explore different styles like anime style...etc.
❤️Pairing with this lora,My favorite base model are below two :
Juggernaut XL
https://civarchive.com/models/133005?modelVersionId=348913
RealVisXL V4.0
https://civarchive.com/models/139562/realvisxl-v40?modelVersionId=344487
Resolutions:
1024*1024
832* 1216
768* 1024
Other SDXL Res will work fine
Recommended Setting for base model Juggernaut XL
Sampler: DPM++ 2M Karras
Steps: 30 to 60 steps
CFG Scale: 7
Resolutions: 1024*1024
For business inquiries, custom models, and consultations, please get in touch at [email protected]
If you need to use this LoRA in your project, please refer to my civitai website: https://civarchive.com/user/lovelydream2023
log
v1: Initial release
If you like this model, please have fun to share your artwork in gallery (please important) and follow me to support us as original creators, and consider giving it a "Positive Review"and a ❤️ and leaving a good review.
Thank you and wish you all happy eyery day.
Enjoy exploring the LoRA!
Kindly note:
Dear all friends, if you want to create the same/similar as the images in the gallery, highly recommend to download this lora into your own computer. Because the gallery's images are with lora weight of 2-4, in contrast, Civitai's online creation is with 1-1.5 of normal lora weight, if over 1.5, then some images will become weird or ugly on Civitai 's generation. I am 100% sure the lora is all good as "tons of" images tests were done before its release. thank you for your attention and kind support to Happy Ai Art Studio.
Disclaimer:
All images generated by the model are created by the users themselves, and the model author cannot control the images created by the users. The model author will not be held responsible for any potential copyright infringement or unsafe images.