Jetstream Sam MGRR | Character LoRA - v1.0
Recommended Prompts
JetstreamSam
cyborg, bodysuit
cyborg, bodysuit, mask
(1boy, solo:1.5), facial hair, male focus, beard, cyborg, scar, scar on face
Highly detailed, High Quality, Masterpiece, beautiful, sharp focus, shiny skin, best quality, perfect anatomy:1.4, high detailed, 8k, highres, lineart, sketch, (masterpiece:1.4), (highest quality:1.4), (sharp focus:1.2), (colored:1.3), (digital painting:1.1), 8k wallpaper, (tight gap:1.2), (1boy, solo:1.5), jetstream sam, facial hair, male focus, beard, cyborg
Recommended Negative Prompts
(badhandv4), (badv4:0.9), (worst quality, low quality:1.4)
Recommended Parameters
samplers
steps
cfg
resolution
other models
Recommended Hires (High-Resolution) Parameters
upscaler
upscale
denoising strength
Tips
Weight can go from 0.5 up to 1. Recommend 0.75 for V2.
Weight can go from 0.7 up to 1. Recommend 0.9 for V1.
ControlNet can help streamline full body shots or more complex poses.
The sword Murasama does not generate properly; expect some inaccuracies.
To use LoRA in Auto1111: Download, copy to stable-diffusion-webui/models/lora, select the LoRA, use the trigger word, and adjust weight if needed.
V2: Jetstream Sam MGRR | Character LoRA
Ohhh? Jetstream got an Update. Very rare.
Refer to the version details for Changelog.
Weight can go from 0.5 up to 1. I recommend 0.75. Works with almost every Style, Concept and Photorealistic Models. Now including his mask. Might be a little unstable, because it's not always covering the ears.
Trigger Word:
JetstreamSam
Outfit Prompt 1:
cyborg, bodysuit
Outfit Prompt 2:
cyborg, bodysuit, mask
V1
Jetstream Sam | MGRR LoRA
Raiden: I'm running late.
Jetstream Sam: Oh don't worry, Raiden. The clown college is closer than you think.
Raiden: Yeah, it's right in front of me.
Sam: Heh heh heh heh. Kill yourself. Your ass will always be cringe, and my ass will always be thicc.
HE IS FINALLY HERE. My first ever trained LoRA now fixed after many attempts. There are not many good images of him, so this LoRA is trained on 30 images. :)
Weight can go from 0.7 up to 1.0. I recommend 0.9. ControlNet will streamline full body shots much easier, or more complex poses. It does NOT work with every Style, every Outfit, Photorealistic Models.
Known problems: The Images are not always accurate. Bear with me, I will update it soon with AI generated images and let it train again for more quality. Also his Sword Murasama is a big problem. Does not generate properly. I will upload some other LoRA's like stances or weapons like Katana or Dual wielding and will try out, how good he is using it.
Trigger Word:
Jetstream Sam
Helping Prompt:
(1boy, solo:1.5), facial hair, male focus, beard, cyborg, scar, scar on face
How to use LoRA's in auto1111:
Download LoRA
Copy the file to
stable-diffusion-webui/models/loraSelect your LoRA and use the triggerword
Make sure to change the weight if needed (by default it's
:1)Have Fun!
Model Details
Model type
Base model
Model version
Model hash
Trained words
Creator
Discussion
Please log in to leave a comment.



