This LoRA is designed to push Krea 2 toward more convincing amateur photography, natural skin texture, and less polished image generation.
It is a concept slider rather than a traditional style LoRA. It does not teach the model a new visual style. Instead, it reinforces details and photographic characteristics that are already present in Krea 2's learned knowledge.
Recommended Strength
Recommended range:
1.0 to 2.0
My current preferred strength is:
1.5
This is a very strong LoRA, so values above 2.0 may produce exaggerated results.
Color Correction
The raw output can be heavily oversaturated.
This is a side effect of the positive and negative concept prompts used during training. Removing the relevant color-related concepts significantly weakened the LoRA.
For best results, use a color correction node after VAE Decode.
My current setup uses the ComfyUI Color Correct node with saturation set between:
-15 and -20
Node I'm using: https://github.com/EllangoK/ComfyUI-post-processing-nodes
You may also need to reduce contrast, temperature, or other values depending on the image.
The example images have been color corrected. Do not expect the raw LoRA output to match them without a similar correction step.
Sampler Setup
I currently use a two-pass ClownSharkSampler workflow.
First Pass
Sampler:
eulerScheduler:
betaSteps:
12
Second Pass
Sampler:
res4s_munthe-kassScheduler:
kl_optimalSteps:
3
My workflow is almost identical to the setup explained in this post:
Notes
Results depend heavily on the base prompt and sampler configuration.
Color correction is strongly recommended.
The LoRA can affect saturation and overall color balance.
It works best when the prompt already describes natural or amateur photography.
This is an experimental release and may not behave consistently across every subject, composition, or aspect ratio.
More detailed explanatory post about this LoRA: https://www.reddit.com/r/comfyui/comments/1v2ycn5/elusarcas_krea_2_smartphone_photography_slider/
Description
FAQ
Comments (18)
Great work!
Looks great, doesn't stack well against character LoRas tho
If you are using the workflow I mentioned with 2 pass sampling, create another model load + lora load pipeline to it. use your character lora here but don't use this lora there. set your character lora strength to 1.5 or close to your first pass lora value. I just gave it a try and it works
@reverentelusarca do you mean an extra model+chara lora pipeline to the second sampler?
@lh983646780717 like this: https://imgur.com/a/8Rx6pBK
Great💕👍🏻
效果惊艳
Saw your reddit post!
Need to check this out as the examples are top notch (I like the pizza and the selfie one) and I like how it keeps everything consistent in the scene.
Dude, THANK YOU! After some extensive testing this gotta be the best LoRA for Krea so far, it gives the subjects very good facial expressions, skin texture AND the lighting is so close to real it is insane lol, plus locations look real, buildings cars look real, clothes look real, everything looks real! your LoRA is just...perfect!
Not just that, your LoRA seen to act as a 'Unlocker' as well, it gets age so well, old people and young people, wrinkles look amazing (see 80 year old).
Thanks again!
Note: The saturation is a plus to me, I enjoy more 'lively' images, currently playing with the LoRA at 0.66
Thank you for your feedback! Let me know if you notice anything off or nice to have for future.
The best 'amateur' Lora so far, and its light in weight. Thanks a bunch! ♥
I couldn't find a saturation node among the default built-in nodes in COMFYUI; there are only brightness and contrast nodes.
My bad, I'll update the post, meanwhile this is the node you can use: https://github.com/EllangoK/ComfyUI-post-processing-nodes
@reverentelusarca THX~
Comfy has a native Blueprint (a.k.a. subgraph) called Color Adjustment, where you can adjust temperature, tint, saturation and vibrance.
@Nexdoor THX~
hey bro, But please share with us the prompts and settings used in your LoRa example images. This way we can achieve better results. Please.
nothing special. they are very simple prompts without giving any photographic details, such as " middle-aged Caucasian man with a bald head and beard, wearing a dark graphic t-shirt and athletic shorts, grilling meat on a stone outdoor kitchen at night, flames rising from the grill, holding tongs, standing in the right third of the frame, medium shot from slightly below eye level, handheld casual angle with slight tilt, foreground includes grill grates, tongs, and spatula resting on counter edge, partially cropped by left frame, warm ambient lighting under range hood, background shows awning and night sky through glass railing"