This LoRA is designed to produce melancholic, cinematic imagery with a strong focus on realism, atmosphere, and emotional restraint. It emphasizes carefully composed scenes, subdued color palettes, and controlled cinematic lighting. The overall look aims for a grounded, highly realistic appearance, often featuring a very shallow depth of field and pronounced background blur, while still allowing images to feel contemplative and emotionally charged.
Recommendation:
Trigger word: zidiusArt
Create images directly at a resolution of around 1328 × 1944 px.
Recommended LoRA strength: start at 0.5 (do not go above 1.0).
Note:
To achieve the intended style, a certain amount of film grain is necessary. For this reason, grain was deliberately included during training. This can sometimes cause issues during upscaling. If you plan to upscale images further, applying some form of denoising as an intermediate step is recommended.
I’m very curious and excited to see what kind of images you will create with this LoRA :)
Description
FAQ
Comments (33)
can u plz add ur workflow and embedd ur settings 😍
My workflow is a total mess at the moment 😄
It’s basically a roughly modified version of my custom Flux workflow – and it looks exactly as bad as it sounds 😅
Most of the stuff isn’t working correctly… or isn’t working at all.
But if you’re looking for something similar, I can highly recommend this one: https://civitai.com/models/2179273/z-imgtxt-to-img
It’s basically an efficient and cleaned-up version of my workflow, not too complicated, and it supports an additional detailer and upscale option to play around with.
(To be honest though, I’m still struggling a bit with the detailer and upscaler)
Other settings are visible in the example images (9 steps, euler, 1.0 CFG; ... nothing special 😉).
@zidius thank you
OMG I love this LoRA... great work!!!
Thank you very much, I really appreciate it 😊
And thanks for the buzz 👍
This lora looks familiar as of Clear_note SDXL lora, Epic stuff, gonna make some images hehe
THIS IS ONE OF THE BEST LORA, JUST INSANE OUTPUT❤️🙏
The pictures turn out very interesting, thank you for the model.
It's similar like Kling 3 image results or you can say better. It's too cinematic. People who want to create stunning cinematic sequence should try this. The only thing is left is dedicated workflow. You can use recommended workflow but It seems not that perfect and with full potential of this lora. you can't resize as this lora recommended. and other things as well.
LORA, created with taste and style. Amazing work 👍 Thank you so much, zidius!
Thank you very much, I really appreciate it 😊
This lora is easily in my top 3 ZiT models. Awesome creation!
Agreed. Just out of curiosity, what are other ZIT LoRAs you think are worth it?
@Nexdoor They definitely have very different purposes, but these are some of my favorites outside of this one:
https://civitai.com/models/2318236?modelVersionId=2608073
https://civitai.com/models/2329053?modelVersionId=2619939
https://civitai.com/models/2234266?modelVersionId=2515203
The detail slider set at 1.3-1.5 is really great for darker themes. It also works really well in conjunction with other loras. I use it with Melancholy a lot.
@Unseelie_AI I didn't know about Purple Grainy, thanks for sharing!
would you mind to post your Lora on tensor art? thanks
That's really cool.
Thank you, really appreciate it!
@Joe_BakerAI Can you please share some of those? :)
@nexyll Actually, my last 20 images use it, but here are some examples. https://civitai.com/images/132030466
@Joe_BakerAI damn nice
Perfect work! Just a question - I noticed that this Lora generates a very blurry background. How can I make the background sharper
@mistania The blurry / out-of-focus background is actually part of the LoRA's style. All of the training images were intentionally created with a strong bokeh and shallow depth of field, so the model learned this as a characteristic feature.
Because of that, it's difficult to get a consistently sharp background while using this LoRA. You can sometimes reduce the effect by lowering the LoRA weight or adjusting your prompt, but the overall look will always tend towards a soft, out-of-focus background.
Awesome Lora! The aesthetics are incredible. I can really feel how hard it is to write prompts. Could you share some tips on writing prompts?
I absolutely LOVE this Lora! I use it on many of my images. Do you have any possible plans to make a version for Krea 2?
Thank you, I'm really glad you like it! If CivitAI adds support for LoRA training for Krea 2, I'll definitely look into it.
@zidius THANK YOU!!!!!
@WikkedAI ♥️
@zidius yessss thank youuuuuu :D