This is the ultimate Hypernetwork for Henry Cavill with 18k steps, 160 images. This Hypernet designed to be flexible and mixed into prompts instead of having a dedicated checkpoint.
***Remember to adjust the weight of the hypernet from 1 downwards (at max 0.30 and as low as 0.10) if you see artifacts/issues***
RECOMMENDED SETTINGS SD 1.5:
Steps: 60
CFG: 3
Prompt Trigger Words: "<hypernet:hyper1-5hnry-18000:0.25> a photo of henry cavill "
Tip#1: Try to use the above Steps and CFG settings first and tweak from there. I think 50 steps may be the minimum acceptable quality. CFG 3 is minimum and after 5, will be a hit or a miss. But you can try other settings to fit your needs.
Tip#2: Using prompt "a luxury modern interior design" works best to trigger the look, then add your own prompt to spice up the picture! Remember to adjust weights as needed!
Tip#3: If you use "RESTORE FACES" option, it will automatically turn Henry's eyes black. I prefer to not use this feature for this, but if you must use it, then you'll need to add "(blue eyes)" to the prompt perhaps with heavy weight to emphasize the blue eyes.
As a favor, please let me know your thoughts, feedback and give a review rating. I would truly appreciate it!
This model was based on a 3D model of a girl working in armor. A mix of science fiction and fantasy, where a girl in a robotic suit and armor. In this case, these are more portraits, and they do not turn out to be full-length. I tried to do experiments with different models. The heroine of this model is Lady Ra, I hope you will like the generated images.
Base model - f222
max steps was - 4000
a close up of a woman wearing a helmet, robort, serious looking, portrait, rococo, looks straight ahead, glowing, realistic fantasy photography, dressed in light armor, soft diffuse lighting, in leather armor, female-focus, imperium, fantasy paladin woman,
<hypernet:LadyRa-3500:0.7> (7 and more)
In full growth(body), of course, they no longer look so elegant, but different generations are quite not bad (^_~)