This is a LoKr-variant of my age-LoRA. Is it better at old people and better at keeping face-details.
But a tiny bit worse on retaining exposure identical at extreme values, as well as scene details.
Not good at anime/cartoons. At least what i found in my testing.
This LoKr can be combined with other LoRAs without any problem.
Recommended range: -3 to 8
Please compare it against the alternatives (including my LoRA-variant) yourself, and please give me feedback if you find any weakness that i can improve i later version.
TIP: This works extremely well in second pass sampler at high values.